Merge branch 'open-webui:main' into main

.dockerignore

@@ -16,4 +16,5 @@ _old
 uploads
 .ipynb_checkpoints
 **/*.db
-_test
+_test
+backend/data/*

.github/workflows/deploy-to-hf-spaces.yml

@@ -28,6 +28,8 @@ jobs:
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
+        with:
+          lfs: true
 
       - name: Remove git history
         run: rm -rf .git
@@ -52,7 +54,9 @@ jobs:
       - name: Set up Git and push to Space
         run: |
           git init --initial-branch=main
+          git lfs install
           git lfs track "*.ttf"
+          git lfs track "*.jpg"
           rm demo.gif
           git add .
           git commit -m "GitHub deploy: ${{ github.sha }}"

CHANGELOG.md

@@ -5,10 +5,81 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.4.2] - 2024-11-20
+
+### Fixed
+
+- **📁 Knowledge Files Visibility Issue**: Resolved the bug preventing individual files in knowledge collections from displaying when referenced with '#'.
+- **🔗 OpenAI Endpoint Prefix**: Fixed the issue where certain OpenAI connections that deviate from the official API spec weren’t working correctly with prefixes.
+- **⚔️ Arena Model Access Control**: Corrected an issue where arena model access control settings were not being saved.
+- **🔧 Usage Capability Selector**: Fixed the broken usage capabilities selector in the model editor.
+
+## [0.4.1] - 2024-11-19
+
+### Added
+
+- **📊 Enhanced Feedback System**: Introduced a detailed 1-10 rating scale for feedback alongside thumbs up/down, preparing for more precise model fine-tuning and improving feedback quality.
+- **ℹ️ Tool Descriptions on Hover**: Easily access tool descriptions by hovering over the message input, providing a smoother workflow with more context when utilizing tools.
+
+### Fixed
+
+- **🗑️ Graceful Handling of Deleted Users**: Resolved an issue where deleted users caused workspace items (models, knowledge, prompts, tools) to fail, ensuring reliable workspace loading.
+- **🔑 API Key Creation**: Fixed an issue preventing users from creating new API keys, restoring secure and seamless API management.
+- **🔗 HTTPS Proxy Fix**: Corrected HTTPS proxy issues affecting the '/api/v1/models/' endpoint, ensuring smoother, uninterrupted model management.
+
+## [0.4.0] - 2024-11-19
+
+### Added
+
+- **👥 User Groups**: You can now create and manage user groups, making user organization seamless.
+- **🔐 Group-Based Access Control**: Set granular access to models, knowledge, prompts, and tools based on user groups, allowing for more controlled and secure environments.
+- **🛠️ Group-Based User Permissions**: Easily manage workspace permissions. Grant users the ability to upload files, delete, edit, or create temporary chats, as well as define their ability to create models, knowledge, prompts, and tools.
+- **🔑 LDAP Support**: Newly introduced LDAP authentication adds robust security and scalability to user management.
+- **🌐 Enhanced OpenAI-Compatible Connections**: Added prefix ID support to avoid model ID clashes, with explicit model ID support for APIs lacking '/models' endpoint support, ensuring smooth operation with custom setups.
+- **🔐 Ollama API Key Support**: Now manage credentials for Ollama when set behind proxies, including the option to utilize prefix ID for proper distinction across multiple Ollama instances.
+- **🔄 Connection Enable/Disable Toggle**: Easily enable or disable individual OpenAI and Ollama connections as needed.
+- **🎨 Redesigned Model Workspace**: Freshly redesigned to improve usability for managing models across users and groups.
+- **🎨 Redesigned Prompt Workspace**: A fresh UI to conveniently organize and manage prompts.
+- **🧩 Sorted Functions Workspace**: Functions are now automatically categorized by type (Action, Filter, Pipe), streamlining management.
+- **💻 Redesigned Collaborative Workspace**: Enhanced support for multiple users contributing to models, knowledge, prompts, or tools, improving collaboration.
+- **🔧 Auto-Selected Tools in Model Editor**: Tools enabled through the model editor are now automatically selected, whereas previously it only gave users the option to enable the tool, reducing manual steps and enhancing efficiency.
+- **🔔 Web Search & Tools Indicator**: A clear indication now shows when web search or tools are active, reducing confusion.
+- **🔑 Toggle API Key Auth**: Tighten security by easily enabling or disabling API key authentication option for Open WebUI.
+- **🗂️ Agentic Retrieval**: Improve RAG accuracy via smart pre-processing of chat history to determine the best queries before retrieval.
+- **📁 Large Text as File Option**: Optionally convert large pasted text into a file upload, keeping the chat interface cleaner.
+- **🗂️ Toggle Citations for Models**: Ability to disable citations has been introduced in the model editor.
+- **🔍 User Settings Search**: Quickly search for settings fields, improving ease of use and navigation.
+- **🗣️ Experimental SpeechT5 TTS**: Local SpeechT5 support added for improved text-to-speech capabilities.
+- **🔄 Unified Reset for Models**: A one-click option has been introduced to reset and remove all models from the Admin Settings.
+- **🛠️ Initial Setup Wizard**: The setup process now explicitly informs users that they are creating an admin account during the first-time setup, ensuring clarity. Previously, users encountered the login page right away without this distinction.
+- **🌐 Enhanced Translations**: Several language translations, including Ukrainian, Norwegian, and Brazilian Portuguese, were refined for better localization.
+
+### Fixed
+
+- **🎥 YouTube Video Attachments**: Fixed issues preventing proper loading and attachment of YouTube videos as files.
+- **🔄 Shared Chat Update**: Corrected issues where shared chats were not updating, improving collaboration consistency.
+- **🔍 DuckDuckGo Rate Limit Fix**: Addressed issues with DuckDuckGo search integration, enhancing search stability and performance when operating within rate limits.
+- **🧾 Citations Relevance Fix**: Adjusted the relevance percentage calculation for citations, so that Open WebUI properly reflect the accuracy of a retrieved document in RAG, ensuring users get clearer insights into sources.
+- **🔑 Jina Search API Key Requirement**: Added the option to input an API key for Jina Search, ensuring smooth functionality as keys are now mandatory.
+
+### Changed
+
+- **🛠️ Functions Moved to Admin Panel**: As Functions operate as advanced plugins, they are now accessible from the Admin Panel instead of the workspace.
+- **🛠️ Manage Ollama Connections**: The "Models" section in Admin Settings has been relocated to Admin Settings > "Connections" > Ollama Connections. You can now manage Ollama instances via a dedicated "Manage Ollama" modal from "Connections", streamlining the setup and configuration of Ollama models.
+- **📊 Base Models in Admin Settings**: Admins can now find all base models, both connections or functions, in the "Models" Admin setting. Global model accessibility can be enabled or disabled here. Models are private by default, requiring explicit permission assignment for user access.
+- **📌 Sticky Model Selection for New Chats**: The model chosen from a previous chat now persists when creating a new chat. If you click "New Chat" again from the new chat page, it will revert to your default model.
+- **🎨 Design Refactoring**: Overall design refinements across the platform have been made, providing a more cohesive and polished user experience.
+
+### Removed
+
+- **📂 Model List Reordering**: Temporarily removed and will be reintroduced in upcoming user group settings improvements.
+- **⚙️ Default Model Setting**: Removed the ability to set a default model for users, will be reintroduced with user group settings in the future.
+
 ## [0.3.35] - 2024-10-26
 
 ### Added
 
+- **🌐 Translation Update**: Added translation labels in the SearchInput and CreateCollection components and updated Brazilian Portuguese translation (pt-BR)
 - **📁 Robust File Handling**: Enhanced file input handling for chat. If the content extraction fails or is empty, users will now receive a clear warning, preventing silent failures and ensuring you always know what's happening with your uploads.
 - **🌍 New Language Support**: Introduced Hungarian translations and updated French translations, expanding the platform's language accessibility for a more global user base.
 

README.md

@@ -21,7 +21,7 @@ Open WebUI is an [extensible](https://github.com/open-webui/pipelines), feature-
 
 - 🤝 **Ollama/OpenAI API Integration**: Effortlessly integrate OpenAI-compatible APIs for versatile conversations alongside Ollama models. Customize the OpenAI API URL to link with **LMStudio, GroqCloud, Mistral, OpenRouter, and more**.
 
-- 🧩 **Pipelines, Open WebUI Plugin Support**: Seamlessly integrate custom logic and Python libraries into Open WebUI using [Pipelines Plugin Framework](https://github.com/open-webui/pipelines). Launch your Pipelines instance, set the OpenAI URL to the Pipelines URL, and explore endless possibilities. [Examples](https://github.com/open-webui/pipelines/tree/main/examples) include **Function Calling**, User **Rate Limiting** to control access, **Usage Monitoring** with tools like Langfuse, **Live Translation with LibreTranslate** for multilingual support, **Toxic Message Filtering** and much more.
+- 🛡️ **Granular Permissions and User Groups**: By allowing administrators to create detailed user roles and permissions, we ensure a secure user environment. This granularity not only enhances security but also allows for customized user experiences, fostering a sense of ownership and responsibility amongst users.
 
 - 📱 **Responsive Design**: Enjoy a seamless experience across Desktop PC, Laptop, and Mobile devices.
 
@@ -37,7 +37,7 @@ Open WebUI is an [extensible](https://github.com/open-webui/pipelines), feature-
 
 - 📚 **Local RAG Integration**: Dive into the future of chat interactions with groundbreaking Retrieval Augmented Generation (RAG) support. This feature seamlessly integrates document interactions into your chat experience. You can load documents directly into the chat or add files to your document library, effortlessly accessing them using the `#` command before a query.
 
-- 🔍 **Web Search for RAG**: Perform web searches using providers like `SearXNG`, `Google PSE`, `Brave Search`, `serpstack`, `serper`, `Serply`, `DuckDuckGo`, `TavilySearch` and `SearchApi` and inject the results directly into your chat experience.
+- 🔍 **Web Search for RAG**: Perform web searches using providers like `SearXNG`, `Google PSE`, `Brave Search`, `serpstack`, `serper`, `Serply`, `DuckDuckGo`, `TavilySearch`, `SearchApi` and `Bing` and inject the results directly into your chat experience.
 
 - 🌐 **Web Browsing Capability**: Seamlessly integrate websites into your chat experience using the `#` command followed by a URL. This feature allows you to incorporate web content directly into your conversations, enhancing the richness and depth of your interactions.
 
@@ -49,6 +49,8 @@ Open WebUI is an [extensible](https://github.com/open-webui/pipelines), feature-
 
 - 🌐🌍 **Multilingual Support**: Experience Open WebUI in your preferred language with our internationalization (i18n) support. Join us in expanding our supported languages! We're actively seeking contributors!
 
+- 🧩 **Pipelines, Open WebUI Plugin Support**: Seamlessly integrate custom logic and Python libraries into Open WebUI using [Pipelines Plugin Framework](https://github.com/open-webui/pipelines). Launch your Pipelines instance, set the OpenAI URL to the Pipelines URL, and explore endless possibilities. [Examples](https://github.com/open-webui/pipelines/tree/main/examples) include **Function Calling**, User **Rate Limiting** to control access, **Usage Monitoring** with tools like Langfuse, **Live Translation with LibreTranslate** for multilingual support, **Toxic Message Filtering** and much more.
+
 - 🌟 **Continuous Updates**: We are committed to improving Open WebUI with regular updates, fixes, and new features.
 
 Want to learn more about Open WebUI's features? Check out our [Open WebUI documentation](https://docs.openwebui.com/features) for a comprehensive overview!
@@ -187,18 +189,6 @@ docker run -d -p 3000:8080 -v open-webui:/app/backend/data --name open-webui --a
 
 Discover upcoming features on our roadmap in the [Open WebUI Documentation](https://docs.openwebui.com/roadmap/).
 
-## Supporters ✨
-
-A big shoutout to our amazing supporters who's helping to make this project possible! 🙏
-
-### Platinum Sponsors 🤍
-
-- We're looking for Sponsors!
-
-### Acknowledgments
-
-Special thanks to [Prof. Lawrence Kim](https://www.lhkim.com/) and [Prof. Nick Vincent](https://www.nickmvincent.com/) for their invaluable support and guidance in shaping this project into a research endeavor. Grateful for your mentorship throughout the journey! 🙌
-
 ## License 📜
 
 This project is licensed under the [MIT License](LICENSE) - see the [LICENSE](LICENSE) file for details. 📄

backend/open_webui/apps/audio/main.py

@@ -32,7 +32,13 @@ from open_webui.config import (
 )
 
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.env import SRC_LOG_LEVELS, DEVICE_TYPE
+from open_webui.env import (
+    ENV,
+    SRC_LOG_LEVELS,
+    DEVICE_TYPE,
+    ENABLE_FORWARD_USER_INFO_HEADERS,
+)
+
 from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile, status
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import FileResponse
@@ -47,7 +53,12 @@ MAX_FILE_SIZE = MAX_FILE_SIZE_MB * 1024 * 1024  # Convert MB to bytes
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["AUDIO"])
 
-app = FastAPI()
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=CORS_ALLOW_ORIGIN,
@@ -74,6 +85,10 @@ app.state.config.TTS_VOICE = AUDIO_TTS_VOICE
 app.state.config.TTS_API_KEY = AUDIO_TTS_API_KEY
 app.state.config.TTS_SPLIT_ON = AUDIO_TTS_SPLIT_ON
 
+
+app.state.speech_synthesiser = None
+app.state.speech_speaker_embeddings_dataset = None
+
 app.state.config.TTS_AZURE_SPEECH_REGION = AUDIO_TTS_AZURE_SPEECH_REGION
 app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT
 
@@ -231,6 +246,21 @@ async def update_audio_config(
     }
 
 
+def load_speech_pipeline():
+    from transformers import pipeline
+    from datasets import load_dataset
+
+    if app.state.speech_synthesiser is None:
+        app.state.speech_synthesiser = pipeline(
+            "text-to-speech", "microsoft/speecht5_tts"
+        )
+
+    if app.state.speech_speaker_embeddings_dataset is None:
+        app.state.speech_speaker_embeddings_dataset = load_dataset(
+            "Matthijs/cmu-arctic-xvectors", split="validation"
+        )
+
+
 @app.post("/speech")
 async def speech(request: Request, user=Depends(get_verified_user)):
     body = await request.body()
@@ -248,6 +278,12 @@ async def speech(request: Request, user=Depends(get_verified_user)):
         headers["Authorization"] = f"Bearer {app.state.config.TTS_OPENAI_API_KEY}"
         headers["Content-Type"] = "application/json"
 
+        if ENABLE_FORWARD_USER_INFO_HEADERS:
+            headers["X-OpenWebUI-User-Name"] = user.name
+            headers["X-OpenWebUI-User-Id"] = user.id
+            headers["X-OpenWebUI-User-Email"] = user.email
+            headers["X-OpenWebUI-User-Role"] = user.role
+
         try:
             body = body.decode("utf-8")
             body = json.loads(body)
@@ -391,6 +427,43 @@ async def speech(request: Request, user=Depends(get_verified_user)):
             raise HTTPException(
                 status_code=500, detail=f"Error synthesizing speech - {response.reason}"
             )
+    elif app.state.config.TTS_ENGINE == "transformers":
+        payload = None
+        try:
+            payload = json.loads(body.decode("utf-8"))
+        except Exception as e:
+            log.exception(e)
+            raise HTTPException(status_code=400, detail="Invalid JSON payload")
+
+        import torch
+        import soundfile as sf
+
+        load_speech_pipeline()
+
+        embeddings_dataset = app.state.speech_speaker_embeddings_dataset
+
+        speaker_index = 6799
+        try:
+            speaker_index = embeddings_dataset["filename"].index(
+                app.state.config.TTS_MODEL
+            )
+        except Exception:
+            pass
+
+        speaker_embedding = torch.tensor(
+            embeddings_dataset[speaker_index]["xvector"]
+        ).unsqueeze(0)
+
+        speech = app.state.speech_synthesiser(
+            payload["input"],
+            forward_params={"speaker_embeddings": speaker_embedding},
+        )
+
+        sf.write(file_path, speech["audio"], samplerate=speech["sampling_rate"])
+        with open(file_body_path, "w") as f:
+            json.dump(json.loads(body.decode("utf-8")), f)
+
+        return FileResponse(file_path)
 
 
 def transcribe(file_path):

backend/open_webui/apps/images/main.py

@@ -35,7 +35,8 @@ from open_webui.config import (
     AppConfig,
 )
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.env import SRC_LOG_LEVELS
+from open_webui.env import ENV, SRC_LOG_LEVELS, ENABLE_FORWARD_USER_INFO_HEADERS
+
 from fastapi import Depends, FastAPI, HTTPException, Request
 from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel
@@ -47,7 +48,12 @@ log.setLevel(SRC_LOG_LEVELS["IMAGES"])
 IMAGE_CACHE_DIR = Path(CACHE_DIR).joinpath("./image/generations/")
 IMAGE_CACHE_DIR.mkdir(parents=True, exist_ok=True)
 
-app = FastAPI()
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=CORS_ALLOW_ORIGIN,
@@ -456,6 +462,12 @@ async def image_generations(
             headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}"
             headers["Content-Type"] = "application/json"
 
+            if ENABLE_FORWARD_USER_INFO_HEADERS:
+                headers["X-OpenWebUI-User-Name"] = user.name
+                headers["X-OpenWebUI-User-Id"] = user.id
+                headers["X-OpenWebUI-User-Email"] = user.email
+                headers["X-OpenWebUI-User-Role"] = user.role
+
             data = {
                 "model": (
                     app.state.config.MODEL

backend/open_webui/apps/ollama/main.py

@@ -13,18 +13,20 @@ import requests
 from open_webui.apps.webui.models.models import Models
 from open_webui.config import (
     CORS_ALLOW_ORIGIN,
-    ENABLE_MODEL_FILTER,
     ENABLE_OLLAMA_API,
-    MODEL_FILTER_LIST,
     OLLAMA_BASE_URLS,
+    OLLAMA_API_CONFIGS,
     UPLOAD_DIR,
     AppConfig,
 )
-from open_webui.env import AIOHTTP_CLIENT_TIMEOUT
+from open_webui.env import (
+    AIOHTTP_CLIENT_TIMEOUT,
+    AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST,
+)
 
 
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.env import SRC_LOG_LEVELS
+from open_webui.env import ENV, SRC_LOG_LEVELS
 from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import StreamingResponse
@@ -41,11 +43,18 @@ from open_webui.utils.payload import (
     apply_model_system_prompt_to_body,
 )
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["OLLAMA"])
 
-app = FastAPI()
+
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=CORS_ALLOW_ORIGIN,
@@ -56,12 +65,9 @@ app.add_middleware(
 
 app.state.config = AppConfig()
 
-app.state.config.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
-app.state.config.MODEL_FILTER_LIST = MODEL_FILTER_LIST
-
 app.state.config.ENABLE_OLLAMA_API = ENABLE_OLLAMA_API
 app.state.config.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS
-app.state.MODELS = {}
+app.state.config.OLLAMA_API_CONFIGS = OLLAMA_API_CONFIGS
 
 
 # TODO: Implement a more intelligent load balancing mechanism for distributing requests among multiple backend instances.
@@ -69,60 +75,98 @@ app.state.MODELS = {}
 # least connections, or least response time for better resource utilization and performance optimization.
 
 
-@app.middleware("http")
-async def check_url(request: Request, call_next):
-    if len(app.state.MODELS) == 0:
-        await get_all_models()
-    else:
-        pass
-
-    response = await call_next(request)
-    return response
-
-
 @app.head("/")
 @app.get("/")
 async def get_status():
     return {"status": True}
 
 
+class ConnectionVerificationForm(BaseModel):
+    url: str
+    key: Optional[str] = None
+
+
+@app.post("/verify")
+async def verify_connection(
+    form_data: ConnectionVerificationForm, user=Depends(get_admin_user)
+):
+    url = form_data.url
+    key = form_data.key
+
+    headers = {}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
+    timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
+    async with aiohttp.ClientSession(timeout=timeout) as session:
+        try:
+            async with session.get(f"{url}/api/version", headers=headers) as r:
+                if r.status != 200:
+                    # Extract response error details if available
+                    error_detail = f"HTTP Error: {r.status}"
+                    res = await r.json()
+                    if "error" in res:
+                        error_detail = f"External Error: {res['error']}"
+                    raise Exception(error_detail)
+
+                response_data = await r.json()
+                return response_data
+
+        except aiohttp.ClientError as e:
+            # ClientError covers all aiohttp requests issues
+            log.exception(f"Client error: {str(e)}")
+            # Handle aiohttp-specific connection issues, timeout etc.
+            raise HTTPException(
+                status_code=500, detail="Open WebUI: Server Connection Error"
+            )
+        except Exception as e:
+            log.exception(f"Unexpected error: {e}")
+            # Generic error handler in case parsing JSON or other steps fail
+            error_detail = f"Unexpected error: {str(e)}"
+            raise HTTPException(status_code=500, detail=error_detail)
+
+
 @app.get("/config")
 async def get_config(user=Depends(get_admin_user)):
-    return {"ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API}
+    return {
+        "ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API,
+        "OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS,
+        "OLLAMA_API_CONFIGS": app.state.config.OLLAMA_API_CONFIGS,
+    }
 
 
 class OllamaConfigForm(BaseModel):
-    enable_ollama_api: Optional[bool] = None
+    ENABLE_OLLAMA_API: Optional[bool] = None
+    OLLAMA_BASE_URLS: list[str]
+    OLLAMA_API_CONFIGS: dict
 
 
 @app.post("/config/update")
 async def update_config(form_data: OllamaConfigForm, user=Depends(get_admin_user)):
-    app.state.config.ENABLE_OLLAMA_API = form_data.enable_ollama_api
-    return {"ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API}
-
-
-@app.get("/urls")
-async def get_ollama_api_urls(user=Depends(get_admin_user)):
-    return {"OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS}
-
+    app.state.config.ENABLE_OLLAMA_API = form_data.ENABLE_OLLAMA_API
+    app.state.config.OLLAMA_BASE_URLS = form_data.OLLAMA_BASE_URLS
 
-class UrlUpdateForm(BaseModel):
-    urls: list[str]
+    app.state.config.OLLAMA_API_CONFIGS = form_data.OLLAMA_API_CONFIGS
 
+    # Remove any extra configs
+    config_urls = app.state.config.OLLAMA_API_CONFIGS.keys()
+    for url in list(app.state.config.OLLAMA_BASE_URLS):
+        if url not in config_urls:
+            app.state.config.OLLAMA_API_CONFIGS.pop(url, None)
 
-@app.post("/urls/update")
-async def update_ollama_api_url(form_data: UrlUpdateForm, user=Depends(get_admin_user)):
-    app.state.config.OLLAMA_BASE_URLS = form_data.urls
+    return {
+        "ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API,
+        "OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS,
+        "OLLAMA_API_CONFIGS": app.state.config.OLLAMA_API_CONFIGS,
+    }
 
-    log.info(f"app.state.config.OLLAMA_BASE_URLS: {app.state.config.OLLAMA_BASE_URLS}")
-    return {"OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS}
 
-
-async def fetch_url(url):
-    timeout = aiohttp.ClientTimeout(total=3)
+async def aiohttp_get(url, key=None):
+    timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
     try:
+        headers = {"Authorization": f"Bearer {key}"} if key else {}
         async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session:
-            async with session.get(url) as response:
+            async with session.get(url, headers=headers) as response:
                 return await response.json()
     except Exception as e:
         # Handle connection error here
@@ -148,10 +192,18 @@ async def post_streaming_url(
         session = aiohttp.ClientSession(
             trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT)
         )
+
+        api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+        key = api_config.get("key", None)
+
+        headers = {"Content-Type": "application/json"}
+        if key:
+            headers["Authorization"] = f"Bearer {key}"
+
         r = await session.post(
             url,
             data=payload,
-            headers={"Content-Type": "application/json"},
+            headers=headers,
         )
         r.raise_for_status()
 
@@ -194,29 +246,62 @@ def merge_models_lists(model_lists):
     for idx, model_list in enumerate(model_lists):
         if model_list is not None:
             for model in model_list:
-                digest = model["digest"]
-                if digest not in merged_models:
+                id = model["model"]
+                if id not in merged_models:
                     model["urls"] = [idx]
-                    merged_models[digest] = model
+                    merged_models[id] = model
                 else:
-                    merged_models[digest]["urls"].append(idx)
+                    merged_models[id]["urls"].append(idx)
 
     return list(merged_models.values())
 
 
 async def get_all_models():
     log.info("get_all_models()")
-
     if app.state.config.ENABLE_OLLAMA_API:
-        tasks = [
-            fetch_url(f"{url}/api/tags") for url in app.state.config.OLLAMA_BASE_URLS
-        ]
+        tasks = []
+        for idx, url in enumerate(app.state.config.OLLAMA_BASE_URLS):
+            if url not in app.state.config.OLLAMA_API_CONFIGS:
+                tasks.append(aiohttp_get(f"{url}/api/tags"))
+            else:
+                api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+                enable = api_config.get("enable", True)
+                key = api_config.get("key", None)
+
+                if enable:
+                    tasks.append(aiohttp_get(f"{url}/api/tags", key))
+                else:
+                    tasks.append(asyncio.ensure_future(asyncio.sleep(0, None)))
+
         responses = await asyncio.gather(*tasks)
 
+        for idx, response in enumerate(responses):
+            if response:
+                url = app.state.config.OLLAMA_BASE_URLS[idx]
+                api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+
+                prefix_id = api_config.get("prefix_id", None)
+                model_ids = api_config.get("model_ids", [])
+
+                if len(model_ids) != 0 and "models" in response:
+                    response["models"] = list(
+                        filter(
+                            lambda model: model["model"] in model_ids,
+                            response["models"],
+                        )
+                    )
+
+                if prefix_id:
+                    for model in response.get("models", []):
+                        model["model"] = f"{prefix_id}.{model['model']}"
+
+        print(responses)
+
         models = {
             "models": merge_models_lists(
                 map(
-                    lambda response: response["models"] if response else None, responses
+                    lambda response: response.get("models", []) if response else None,
+                    responses,
                 )
             )
         }
@@ -224,8 +309,6 @@ async def get_all_models():
     else:
         models = {"models": []}
 
-    app.state.MODELS = {model["model"]: model for model in models["models"]}
-
     return models
 
 
@@ -234,29 +317,25 @@ async def get_all_models():
 async def get_ollama_tags(
     url_idx: Optional[int] = None, user=Depends(get_verified_user)
 ):
+    models = []
     if url_idx is None:
         models = await get_all_models()
-
-        if app.state.config.ENABLE_MODEL_FILTER:
-            if user.role == "user":
-                models["models"] = list(
-                    filter(
-                        lambda model: model["name"]
-                        in app.state.config.MODEL_FILTER_LIST,
-                        models["models"],
-                    )
-                )
-                return models
-        return models
     else:
         url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 
+        api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+        key = api_config.get("key", None)
+
+        headers = {}
+        if key:
+            headers["Authorization"] = f"Bearer {key}"
+
         r = None
         try:
-            r = requests.request(method="GET", url=f"{url}/api/tags")
+            r = requests.request(method="GET", url=f"{url}/api/tags", headers=headers)
             r.raise_for_status()
 
-            return r.json()
+            models = r.json()
         except Exception as e:
             log.exception(e)
             error_detail = "Open WebUI: Server Connection Error"
@@ -273,6 +352,20 @@ async def get_ollama_tags(
                 detail=error_detail,
             )
 
+    if user.role == "user":
+        # Filter models based on user access control
+        filtered_models = []
+        for model in models.get("models", []):
+            model_info = Models.get_model_by_id(model["model"])
+            if model_info:
+                if user.id == model_info.user_id or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                ):
+                    filtered_models.append(model)
+        models["models"] = filtered_models
+
+    return models
+
 
 @app.get("/api/version")
 @app.get("/api/version/{url_idx}")
@@ -281,7 +374,10 @@ async def get_ollama_versions(url_idx: Optional[int] = None):
         if url_idx is None:
             # returns lowest version
             tasks = [
-                fetch_url(f"{url}/api/version")
+                aiohttp_get(
+                    f"{url}/api/version",
+                    app.state.config.OLLAMA_API_CONFIGS.get(url, {}).get("key", None),
+                )
                 for url in app.state.config.OLLAMA_BASE_URLS
             ]
             responses = await asyncio.gather(*tasks)
@@ -361,8 +457,11 @@ async def push_model(
     user=Depends(get_admin_user),
 ):
     if url_idx is None:
-        if form_data.name in app.state.MODELS:
-            url_idx = app.state.MODELS[form_data.name]["urls"][0]
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
+        if form_data.name in models:
+            url_idx = models[form_data.name]["urls"][0]
         else:
             raise HTTPException(
                 status_code=400,
@@ -411,8 +510,11 @@ async def copy_model(
     user=Depends(get_admin_user),
 ):
     if url_idx is None:
-        if form_data.source in app.state.MODELS:
-            url_idx = app.state.MODELS[form_data.source]["urls"][0]
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
+        if form_data.source in models:
+            url_idx = models[form_data.source]["urls"][0]
         else:
             raise HTTPException(
                 status_code=400,
@@ -421,10 +523,18 @@ async def copy_model(
 
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
+
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    key = api_config.get("key", None)
+
+    headers = {"Content-Type": "application/json"}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
     r = requests.request(
         method="POST",
         url=f"{url}/api/copy",
-        headers={"Content-Type": "application/json"},
+        headers=headers,
         data=form_data.model_dump_json(exclude_none=True).encode(),
     )
 
@@ -459,8 +569,11 @@ async def delete_model(
     user=Depends(get_admin_user),
 ):
     if url_idx is None:
-        if form_data.name in app.state.MODELS:
-            url_idx = app.state.MODELS[form_data.name]["urls"][0]
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
+        if form_data.name in models:
+            url_idx = models[form_data.name]["urls"][0]
         else:
             raise HTTPException(
                 status_code=400,
@@ -470,11 +583,18 @@ async def delete_model(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    key = api_config.get("key", None)
+
+    headers = {"Content-Type": "application/json"}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
     r = requests.request(
         method="DELETE",
         url=f"{url}/api/delete",
-        headers={"Content-Type": "application/json"},
         data=form_data.model_dump_json(exclude_none=True).encode(),
+        headers=headers,
     )
     try:
         r.raise_for_status()
@@ -501,20 +621,30 @@ async def delete_model(
 
 @app.post("/api/show")
 async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_user)):
-    if form_data.name not in app.state.MODELS:
+    model_list = await get_all_models()
+    models = {model["model"]: model for model in model_list["models"]}
+
+    if form_data.name not in models:
         raise HTTPException(
             status_code=400,
             detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
         )
 
-    url_idx = random.choice(app.state.MODELS[form_data.name]["urls"])
+    url_idx = random.choice(models[form_data.name]["urls"])
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    key = api_config.get("key", None)
+
+    headers = {"Content-Type": "application/json"}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
     r = requests.request(
         method="POST",
         url=f"{url}/api/show",
-        headers={"Content-Type": "application/json"},
+        headers=headers,
         data=form_data.model_dump_json(exclude_none=True).encode(),
     )
     try:
@@ -570,23 +700,26 @@ async def generate_embeddings(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-    return generate_ollama_embeddings(form_data=form_data, url_idx=url_idx)
+    return await generate_ollama_embeddings(form_data=form_data, url_idx=url_idx)
 
 
-def generate_ollama_embeddings(
+async def generate_ollama_embeddings(
     form_data: GenerateEmbeddingsForm,
     url_idx: Optional[int] = None,
 ):
     log.info(f"generate_ollama_embeddings {form_data}")
 
     if url_idx is None:
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
         model = form_data.model
 
         if ":" not in model:
             model = f"{model}:latest"
 
-        if model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[model]["urls"])
+        if model in models:
+            url_idx = random.choice(models[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
@@ -596,10 +729,17 @@ def generate_ollama_embeddings(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    key = api_config.get("key", None)
+
+    headers = {"Content-Type": "application/json"}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
     r = requests.request(
         method="POST",
         url=f"{url}/api/embeddings",
-        headers={"Content-Type": "application/json"},
+        headers=headers,
         data=form_data.model_dump_json(exclude_none=True).encode(),
     )
     try:
@@ -630,20 +770,23 @@ def generate_ollama_embeddings(
         )
 
 
-def generate_ollama_batch_embeddings(
+async def generate_ollama_batch_embeddings(
     form_data: GenerateEmbedForm,
     url_idx: Optional[int] = None,
 ):
     log.info(f"generate_ollama_batch_embeddings {form_data}")
 
     if url_idx is None:
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
         model = form_data.model
 
         if ":" not in model:
             model = f"{model}:latest"
 
-        if model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[model]["urls"])
+        if model in models:
+            url_idx = random.choice(models[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
@@ -653,10 +796,17 @@ def generate_ollama_batch_embeddings(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    key = api_config.get("key", None)
+
+    headers = {"Content-Type": "application/json"}
+    if key:
+        headers["Authorization"] = f"Bearer {key}"
+
     r = requests.request(
         method="POST",
         url=f"{url}/api/embed",
-        headers={"Content-Type": "application/json"},
+        headers=headers,
         data=form_data.model_dump_json(exclude_none=True).encode(),
     )
     try:
@@ -692,7 +842,7 @@ class GenerateCompletionForm(BaseModel):
     options: Optional[dict] = None
     system: Optional[str] = None
     template: Optional[str] = None
-    context: Optional[str] = None
+    context: Optional[list[int]] = None
     stream: Optional[bool] = True
     raw: Optional[bool] = None
     keep_alive: Optional[Union[int, str]] = None
@@ -706,13 +856,16 @@ async def generate_completion(
     user=Depends(get_verified_user),
 ):
     if url_idx is None:
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
         model = form_data.model
 
         if ":" not in model:
             model = f"{model}:latest"
 
-        if model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[model]["urls"])
+        if model in models:
+            url_idx = random.choice(models[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
@@ -720,6 +873,10 @@ async def generate_completion(
             )
 
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    prefix_id = api_config.get("prefix_id", None)
+    if prefix_id:
+        form_data.model = form_data.model.replace(f"{prefix_id}.", "")
     log.info(f"url: {url}")
 
     return await post_streaming_url(
@@ -743,14 +900,17 @@ class GenerateChatCompletionForm(BaseModel):
     keep_alive: Optional[Union[int, str]] = None
 
 
-def get_ollama_url(url_idx: Optional[int], model: str):
+async def get_ollama_url(url_idx: Optional[int], model: str):
     if url_idx is None:
-        if model not in app.state.MODELS:
+        model_list = await get_all_models()
+        models = {model["model"]: model for model in model_list["models"]}
+
+        if model not in models:
             raise HTTPException(
                 status_code=400,
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(model),
             )
-        url_idx = random.choice(app.state.MODELS[model]["urls"])
+        url_idx = random.choice(models[model]["urls"])
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     return url
 
@@ -768,15 +928,7 @@ async def generate_chat_completion(
     if "metadata" in payload:
         del payload["metadata"]
 
-    model_id = form_data.model
-
-    if not bypass_filter and app.state.config.ENABLE_MODEL_FILTER:
-        if user.role == "user" and model_id not in app.state.config.MODEL_FILTER_LIST:
-            raise HTTPException(
-                status_code=403,
-                detail="Model not found",
-            )
-
+    model_id = payload["model"]
     model_info = Models.get_model_by_id(model_id)
 
     if model_info:
@@ -794,13 +946,37 @@ async def generate_chat_completion(
             )
             payload = apply_model_system_prompt_to_body(params, payload, user)
 
+        # Check if user has access to the model
+        if not bypass_filter and user.role == "user":
+            if not (
+                user.id == model_info.user_id
+                or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                )
+            ):
+                raise HTTPException(
+                    status_code=403,
+                    detail="Model not found",
+                )
+    elif not bypass_filter:
+        if user.role != "admin":
+            raise HTTPException(
+                status_code=403,
+                detail="Model not found",
+            )
+
     if ":" not in payload["model"]:
         payload["model"] = f"{payload['model']}:latest"
 
-    url = get_ollama_url(url_idx, payload["model"])
+    url = await get_ollama_url(url_idx, payload["model"])
     log.info(f"url: {url}")
     log.debug(f"generate_chat_completion() - 2.payload = {payload}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    prefix_id = api_config.get("prefix_id", None)
+    if prefix_id:
+        payload["model"] = payload["model"].replace(f"{prefix_id}.", "")
+
     return await post_streaming_url(
         f"{url}/api/chat",
         json.dumps(payload),
@@ -817,7 +993,7 @@ class OpenAIChatMessageContent(BaseModel):
 
 class OpenAIChatMessage(BaseModel):
     role: str
-    content: Union[str, OpenAIChatMessageContent]
+    content: Union[str, list[OpenAIChatMessageContent]]
 
     model_config = ConfigDict(extra="allow")
 
@@ -836,22 +1012,24 @@ async def generate_openai_chat_completion(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-    completion_form = OpenAIChatCompletionForm(**form_data)
+    try:
+        completion_form = OpenAIChatCompletionForm(**form_data)
+    except Exception as e:
+        log.exception(e)
+        raise HTTPException(
+            status_code=400,
+            detail=str(e),
+        )
+
     payload = {**completion_form.model_dump(exclude_none=True, exclude=["metadata"])}
     if "metadata" in payload:
         del payload["metadata"]
 
     model_id = completion_form.model
-
-    if app.state.config.ENABLE_MODEL_FILTER:
-        if user.role == "user" and model_id not in app.state.config.MODEL_FILTER_LIST:
-            raise HTTPException(
-                status_code=403,
-                detail="Model not found",
-            )
+    if ":" not in model_id:
+        model_id = f"{model_id}:latest"
 
     model_info = Models.get_model_by_id(model_id)
-
     if model_info:
         if model_info.base_model_id:
             payload["model"] = model_info.base_model_id
@@ -862,12 +1040,36 @@ async def generate_openai_chat_completion(
             payload = apply_model_params_to_body_openai(params, payload)
             payload = apply_model_system_prompt_to_body(params, payload, user)
 
+        # Check if user has access to the model
+        if user.role == "user":
+            if not (
+                user.id == model_info.user_id
+                or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                )
+            ):
+                raise HTTPException(
+                    status_code=403,
+                    detail="Model not found",
+                )
+    else:
+        if user.role != "admin":
+            raise HTTPException(
+                status_code=403,
+                detail="Model not found",
+            )
+
     if ":" not in payload["model"]:
         payload["model"] = f"{payload['model']}:latest"
 
-    url = get_ollama_url(url_idx, payload["model"])
+    url = await get_ollama_url(url_idx, payload["model"])
     log.info(f"url: {url}")
 
+    api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {})
+    prefix_id = api_config.get("prefix_id", None)
+    if prefix_id:
+        payload["model"] = payload["model"].replace(f"{prefix_id}.", "")
+
     return await post_streaming_url(
         f"{url}/v1/chat/completions",
         json.dumps(payload),
@@ -881,21 +1083,29 @@ async def get_openai_models(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
+
+    models = []
     if url_idx is None:
-        models = await get_all_models()
+        model_list = await get_all_models()
+        models = [
+            {
+                "id": model["model"],
+                "object": "model",
+                "created": int(time.time()),
+                "owned_by": "openai",
+            }
+            for model in model_list["models"]
+        ]
 
-        if app.state.config.ENABLE_MODEL_FILTER:
-            if user.role == "user":
-                models["models"] = list(
-                    filter(
-                        lambda model: model["name"]
-                        in app.state.config.MODEL_FILTER_LIST,
-                        models["models"],
-                    )
-                )
+    else:
+        url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+        try:
+            r = requests.request(method="GET", url=f"{url}/api/tags")
+            r.raise_for_status()
+
+            model_list = r.json()
 
-        return {
-            "data": [
+            models = [
                 {
                     "id": model["model"],
                     "object": "model",
@@ -903,31 +1113,7 @@ async def get_openai_models(
                     "owned_by": "openai",
                 }
                 for model in models["models"]
-            ],
-            "object": "list",
-        }
-
-    else:
-        url = app.state.config.OLLAMA_BASE_URLS[url_idx]
-        try:
-            r = requests.request(method="GET", url=f"{url}/api/tags")
-            r.raise_for_status()
-
-            models = r.json()
-
-            return {
-                "data": [
-                    {
-                        "id": model["model"],
-                        "object": "model",
-                        "created": int(time.time()),
-                        "owned_by": "openai",
-                    }
-                    for model in models["models"]
-                ],
-                "object": "list",
-            }
-
+            ]
         except Exception as e:
             log.exception(e)
             error_detail = "Open WebUI: Server Connection Error"
@@ -944,6 +1130,23 @@ async def get_openai_models(
                 detail=error_detail,
             )
 
+    if user.role == "user":
+        # Filter models based on user access control
+        filtered_models = []
+        for model in models:
+            model_info = Models.get_model_by_id(model["id"])
+            if model_info:
+                if user.id == model_info.user_id or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                ):
+                    filtered_models.append(model)
+        models = filtered_models
+
+    return {
+        "data": models,
+        "object": "list",
+    }
+
 
 class UrlForm(BaseModel):
     url: str

backend/open_webui/apps/openai/main.py

@@ -11,20 +11,20 @@ from open_webui.apps.webui.models.models import Models
 from open_webui.config import (
     CACHE_DIR,
     CORS_ALLOW_ORIGIN,
-    ENABLE_MODEL_FILTER,
     ENABLE_OPENAI_API,
-    MODEL_FILTER_LIST,
     OPENAI_API_BASE_URLS,
     OPENAI_API_KEYS,
+    OPENAI_API_CONFIGS,
     AppConfig,
 )
 from open_webui.env import (
     AIOHTTP_CLIENT_TIMEOUT,
     AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST,
+    ENABLE_FORWARD_USER_INFO_HEADERS,
 )
 
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.env import SRC_LOG_LEVELS
+from open_webui.env import ENV, SRC_LOG_LEVELS
 from fastapi import Depends, FastAPI, HTTPException, Request
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import FileResponse, StreamingResponse
@@ -37,11 +37,20 @@ from open_webui.utils.payload import (
 )
 
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access
+
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["OPENAI"])
 
-app = FastAPI()
+
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
+
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=CORS_ALLOW_ORIGIN,
@@ -52,69 +61,66 @@ app.add_middleware(
 
 app.state.config = AppConfig()
 
-app.state.config.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
-app.state.config.MODEL_FILTER_LIST = MODEL_FILTER_LIST
-
 app.state.config.ENABLE_OPENAI_API = ENABLE_OPENAI_API
 app.state.config.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS
 app.state.config.OPENAI_API_KEYS = OPENAI_API_KEYS
-
-app.state.MODELS = {}
-
-
-@app.middleware("http")
-async def check_url(request: Request, call_next):
-    if len(app.state.MODELS) == 0:
-        await get_all_models()
-
-    response = await call_next(request)
-    return response
+app.state.config.OPENAI_API_CONFIGS = OPENAI_API_CONFIGS
 
 
 @app.get("/config")
 async def get_config(user=Depends(get_admin_user)):
-    return {"ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API}
+    return {
+        "ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API,
+        "OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS,
+        "OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS,
+        "OPENAI_API_CONFIGS": app.state.config.OPENAI_API_CONFIGS,
+    }
 
 
 class OpenAIConfigForm(BaseModel):
-    enable_openai_api: Optional[bool] = None
+    ENABLE_OPENAI_API: Optional[bool] = None
+    OPENAI_API_BASE_URLS: list[str]
+    OPENAI_API_KEYS: list[str]
+    OPENAI_API_CONFIGS: dict
 
 
 @app.post("/config/update")
 async def update_config(form_data: OpenAIConfigForm, user=Depends(get_admin_user)):
-    app.state.config.ENABLE_OPENAI_API = form_data.enable_openai_api
-    return {"ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API}
-
-
-class UrlsUpdateForm(BaseModel):
-    urls: list[str]
+    app.state.config.ENABLE_OPENAI_API = form_data.ENABLE_OPENAI_API
 
+    app.state.config.OPENAI_API_BASE_URLS = form_data.OPENAI_API_BASE_URLS
+    app.state.config.OPENAI_API_KEYS = form_data.OPENAI_API_KEYS
 
-class KeysUpdateForm(BaseModel):
-    keys: list[str]
-
-
-@app.get("/urls")
-async def get_openai_urls(user=Depends(get_admin_user)):
-    return {"OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS}
-
-
-@app.post("/urls/update")
-async def update_openai_urls(form_data: UrlsUpdateForm, user=Depends(get_admin_user)):
-    await get_all_models()
-    app.state.config.OPENAI_API_BASE_URLS = form_data.urls
-    return {"OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS}
-
+    # Check if API KEYS length is same than API URLS length
+    if len(app.state.config.OPENAI_API_KEYS) != len(
+        app.state.config.OPENAI_API_BASE_URLS
+    ):
+        if len(app.state.config.OPENAI_API_KEYS) > len(
+            app.state.config.OPENAI_API_BASE_URLS
+        ):
+            app.state.config.OPENAI_API_KEYS = app.state.config.OPENAI_API_KEYS[
+                : len(app.state.config.OPENAI_API_BASE_URLS)
+            ]
+        else:
+            app.state.config.OPENAI_API_KEYS += [""] * (
+                len(app.state.config.OPENAI_API_BASE_URLS)
+                - len(app.state.config.OPENAI_API_KEYS)
+            )
 
-@app.get("/keys")
-async def get_openai_keys(user=Depends(get_admin_user)):
-    return {"OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS}
+    app.state.config.OPENAI_API_CONFIGS = form_data.OPENAI_API_CONFIGS
 
+    # Remove any extra configs
+    config_urls = app.state.config.OPENAI_API_CONFIGS.keys()
+    for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS):
+        if url not in config_urls:
+            app.state.config.OPENAI_API_CONFIGS.pop(url, None)
 
-@app.post("/keys/update")
-async def update_openai_key(form_data: KeysUpdateForm, user=Depends(get_admin_user)):
-    app.state.config.OPENAI_API_KEYS = form_data.keys
-    return {"OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS}
+    return {
+        "ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API,
+        "OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS,
+        "OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS,
+        "OPENAI_API_CONFIGS": app.state.config.OPENAI_API_CONFIGS,
+    }
 
 
 @app.post("/audio/speech")
@@ -140,6 +146,11 @@ async def speech(request: Request, user=Depends(get_verified_user)):
         if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]:
             headers["HTTP-Referer"] = "https://openwebui.com/"
             headers["X-Title"] = "Open WebUI"
+        if ENABLE_FORWARD_USER_INFO_HEADERS:
+            headers["X-OpenWebUI-User-Name"] = user.name
+            headers["X-OpenWebUI-User-Id"] = user.id
+            headers["X-OpenWebUI-User-Email"] = user.email
+            headers["X-OpenWebUI-User-Role"] = user.role
         r = None
         try:
             r = requests.post(
@@ -181,10 +192,10 @@ async def speech(request: Request, user=Depends(get_verified_user)):
         raise HTTPException(status_code=401, detail=ERROR_MESSAGES.OPENAI_NOT_FOUND)
 
 
-async def fetch_url(url, key):
+async def aiohttp_get(url, key=None):
     timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
     try:
-        headers = {"Authorization": f"Bearer {key}"}
+        headers = {"Authorization": f"Bearer {key}"} if key else {}
         async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session:
             async with session.get(url, headers=headers) as response:
                 return await response.json()
@@ -239,12 +250,8 @@ def merge_models_lists(model_lists):
     return merged_list
 
 
-def is_openai_api_disabled():
-    return not app.state.config.ENABLE_OPENAI_API
-
-
-async def get_all_models_raw() -> list:
-    if is_openai_api_disabled():
+async def get_all_models_responses() -> list:
+    if not app.state.config.ENABLE_OPENAI_API:
         return []
 
     # Check if API KEYS length is same than API URLS length
@@ -260,33 +267,69 @@ async def get_all_models_raw() -> list:
         else:
             app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys)
 
-    tasks = [
-        fetch_url(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx])
-        for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS)
-    ]
+    tasks = []
+    for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS):
+        if url not in app.state.config.OPENAI_API_CONFIGS:
+            tasks.append(
+                aiohttp_get(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx])
+            )
+        else:
+            api_config = app.state.config.OPENAI_API_CONFIGS.get(url, {})
+
+            enable = api_config.get("enable", True)
+            model_ids = api_config.get("model_ids", [])
+
+            if enable:
+                if len(model_ids) == 0:
+                    tasks.append(
+                        aiohttp_get(
+                            f"{url}/models", app.state.config.OPENAI_API_KEYS[idx]
+                        )
+                    )
+                else:
+                    model_list = {
+                        "object": "list",
+                        "data": [
+                            {
+                                "id": model_id,
+                                "name": model_id,
+                                "owned_by": "openai",
+                                "openai": {"id": model_id},
+                                "urlIdx": idx,
+                            }
+                            for model_id in model_ids
+                        ],
+                    }
+
+                    tasks.append(asyncio.ensure_future(asyncio.sleep(0, model_list)))
 
     responses = await asyncio.gather(*tasks)
-    log.debug(f"get_all_models:responses() {responses}")
 
-    return responses
+    for idx, response in enumerate(responses):
+        if response:
+            url = app.state.config.OPENAI_API_BASE_URLS[idx]
+            api_config = app.state.config.OPENAI_API_CONFIGS.get(url, {})
 
+            prefix_id = api_config.get("prefix_id", None)
 
-@overload
-async def get_all_models(raw: Literal[True]) -> list: ...
+            if prefix_id:
+                for model in (
+                    response if isinstance(response, list) else response.get("data", [])
+                ):
+                    model["id"] = f"{prefix_id}.{model['id']}"
 
+    log.debug(f"get_all_models:responses() {responses}")
 
-@overload
-async def get_all_models(raw: Literal[False] = False) -> dict[str, list]: ...
+    return responses
 
 
-async def get_all_models(raw=False) -> dict[str, list] | list:
+async def get_all_models() -> dict[str, list]:
     log.info("get_all_models()")
-    if is_openai_api_disabled():
-        return [] if raw else {"data": []}
 
-    responses = await get_all_models_raw()
-    if raw:
-        return responses
+    if not app.state.config.ENABLE_OPENAI_API:
+        return {"data": []}
+
+    responses = await get_all_models_responses()
 
     def extract_data(response):
         if response and "data" in response:
@@ -296,9 +339,7 @@ async def get_all_models(raw=False) -> dict[str, list] | list:
         return None
 
     models = {"data": merge_models_lists(map(extract_data, responses))}
-
     log.debug(f"models: {models}")
-    app.state.MODELS = {model["id"]: model for model in models["data"]}
 
     return models
 
@@ -306,18 +347,12 @@ async def get_all_models(raw=False) -> dict[str, list] | list:
 @app.get("/models")
 @app.get("/models/{url_idx}")
 async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_user)):
+    models = {
+        "data": [],
+    }
+
     if url_idx is None:
         models = await get_all_models()
-        if app.state.config.ENABLE_MODEL_FILTER:
-            if user.role == "user":
-                models["data"] = list(
-                    filter(
-                        lambda model: model["id"] in app.state.config.MODEL_FILTER_LIST,
-                        models["data"],
-                    )
-                )
-                return models
-        return models
     else:
         url = app.state.config.OPENAI_API_BASE_URLS[url_idx]
         key = app.state.config.OPENAI_API_KEYS[url_idx]
@@ -326,56 +361,126 @@ async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_us
         headers["Authorization"] = f"Bearer {key}"
         headers["Content-Type"] = "application/json"
 
+        if ENABLE_FORWARD_USER_INFO_HEADERS:
+            headers["X-OpenWebUI-User-Name"] = user.name
+            headers["X-OpenWebUI-User-Id"] = user.id
+            headers["X-OpenWebUI-User-Email"] = user.email
+            headers["X-OpenWebUI-User-Role"] = user.role
+
         r = None
 
-        try:
-            r = requests.request(method="GET", url=f"{url}/models", headers=headers)
-            r.raise_for_status()
+        timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
+        async with aiohttp.ClientSession(timeout=timeout) as session:
+            try:
+                async with session.get(f"{url}/models", headers=headers) as r:
+                    if r.status != 200:
+                        # Extract response error details if available
+                        error_detail = f"HTTP Error: {r.status}"
+                        res = await r.json()
+                        if "error" in res:
+                            error_detail = f"External Error: {res['error']}"
+                        raise Exception(error_detail)
+
+                    response_data = await r.json()
+
+                    # Check if we're calling OpenAI API based on the URL
+                    if "api.openai.com" in url:
+                        # Filter models according to the specified conditions
+                        response_data["data"] = [
+                            model
+                            for model in response_data.get("data", [])
+                            if not any(
+                                name in model["id"]
+                                for name in [
+                                    "babbage",
+                                    "dall-e",
+                                    "davinci",
+                                    "embedding",
+                                    "tts",
+                                    "whisper",
+                                ]
+                            )
+                        ]
 
-            response_data = r.json()
+                    models = response_data
+            except aiohttp.ClientError as e:
+                # ClientError covers all aiohttp requests issues
+                log.exception(f"Client error: {str(e)}")
+                # Handle aiohttp-specific connection issues, timeout etc.
+                raise HTTPException(
+                    status_code=500, detail="Open WebUI: Server Connection Error"
+                )
+            except Exception as e:
+                log.exception(f"Unexpected error: {e}")
+                # Generic error handler in case parsing JSON or other steps fail
+                error_detail = f"Unexpected error: {str(e)}"
+                raise HTTPException(status_code=500, detail=error_detail)
+
+    if user.role == "user":
+        # Filter models based on user access control
+        filtered_models = []
+        for model in models.get("data", []):
+            model_info = Models.get_model_by_id(model["id"])
+            if model_info:
+                if user.id == model_info.user_id or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                ):
+                    filtered_models.append(model)
+        models["data"] = filtered_models
 
-            if "api.openai.com" in url:
-                # Filter the response data
-                response_data["data"] = [
-                    model
-                    for model in response_data["data"]
-                    if not any(
-                        name in model["id"]
-                        for name in [
-                            "babbage",
-                            "dall-e",
-                            "davinci",
-                            "embedding",
-                            "tts",
-                            "whisper",
-                        ]
-                    )
-                ]
+    return models
 
-            return response_data
-        except Exception as e:
-            log.exception(e)
-            error_detail = "Open WebUI: Server Connection Error"
-            if r is not None:
-                try:
-                    res = r.json()
+
+class ConnectionVerificationForm(BaseModel):
+    url: str
+    key: str
+
+
+@app.post("/verify")
+async def verify_connection(
+    form_data: ConnectionVerificationForm, user=Depends(get_admin_user)
+):
+    url = form_data.url
+    key = form_data.key
+
+    headers = {}
+    headers["Authorization"] = f"Bearer {key}"
+    headers["Content-Type"] = "application/json"
+
+    timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
+    async with aiohttp.ClientSession(timeout=timeout) as session:
+        try:
+            async with session.get(f"{url}/models", headers=headers) as r:
+                if r.status != 200:
+                    # Extract response error details if available
+                    error_detail = f"HTTP Error: {r.status}"
+                    res = await r.json()
                     if "error" in res:
-                        error_detail = f"External: {res['error']}"
-                except Exception:
-                    error_detail = f"External: {e}"
+                        error_detail = f"External Error: {res['error']}"
+                    raise Exception(error_detail)
+
+                response_data = await r.json()
+                return response_data
 
+        except aiohttp.ClientError as e:
+            # ClientError covers all aiohttp requests issues
+            log.exception(f"Client error: {str(e)}")
+            # Handle aiohttp-specific connection issues, timeout etc.
             raise HTTPException(
-                status_code=r.status_code if r else 500,
-                detail=error_detail,
+                status_code=500, detail="Open WebUI: Server Connection Error"
             )
+        except Exception as e:
+            log.exception(f"Unexpected error: {e}")
+            # Generic error handler in case parsing JSON or other steps fail
+            error_detail = f"Unexpected error: {str(e)}"
+            raise HTTPException(status_code=500, detail=error_detail)
 
 
 @app.post("/chat/completions")
-@app.post("/chat/completions/{url_idx}")
 async def generate_chat_completion(
     form_data: dict,
-    url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
+    bypass_filter: Optional[bool] = False,
 ):
     idx = 0
     payload = {**form_data}
@@ -386,6 +491,7 @@ async def generate_chat_completion(
     model_id = form_data.get("model")
     model_info = Models.get_model_by_id(model_id)
 
+    # Check model info and override the payload
     if model_info:
         if model_info.base_model_id:
             payload["model"] = model_info.base_model_id
@@ -394,9 +500,52 @@ async def generate_chat_completion(
         payload = apply_model_params_to_body_openai(params, payload)
         payload = apply_model_system_prompt_to_body(params, payload, user)
 
-    model = app.state.MODELS[payload.get("model")]
-    idx = model["urlIdx"]
+        # Check if user has access to the model
+        if not bypass_filter and user.role == "user":
+            if not (
+                user.id == model_info.user_id
+                or has_access(
+                    user.id, type="read", access_control=model_info.access_control
+                )
+            ):
+                raise HTTPException(
+                    status_code=403,
+                    detail="Model not found",
+                )
+    elif not bypass_filter:
+        if user.role != "admin":
+            raise HTTPException(
+                status_code=403,
+                detail="Model not found",
+            )
+
+    # Attemp to get urlIdx from the model
+    models = await get_all_models()
 
+    # Find the model from the list
+    model = next(
+        (model for model in models["data"] if model["id"] == payload.get("model")),
+        None,
+    )
+
+    if model:
+        idx = model["urlIdx"]
+    else:
+        raise HTTPException(
+            status_code=404,
+            detail="Model not found",
+        )
+
+    # Get the API config for the model
+    api_config = app.state.config.OPENAI_API_CONFIGS.get(
+        app.state.config.OPENAI_API_BASE_URLS[idx], {}
+    )
+    prefix_id = api_config.get("prefix_id", None)
+
+    if prefix_id:
+        payload["model"] = payload["model"].replace(f"{prefix_id}.", "")
+
+    # Add user info to the payload if the model is a pipeline
     if "pipeline" in model and model.get("pipeline"):
         payload["user"] = {
             "name": user.name,
@@ -407,8 +556,9 @@ async def generate_chat_completion(
 
     url = app.state.config.OPENAI_API_BASE_URLS[idx]
     key = app.state.config.OPENAI_API_KEYS[idx]
-    is_o1 = payload["model"].lower().startswith("o1-")
 
+    # Fix: O1 does not support the "max_tokens" parameter, Modify "max_tokens" to "max_completion_tokens"
+    is_o1 = payload["model"].lower().startswith("o1-")
     # Change max_completion_tokens to max_tokens (Backward compatible)
     if "api.openai.com" not in url and not is_o1:
         if "max_completion_tokens" in payload:
@@ -437,6 +587,11 @@ async def generate_chat_completion(
     if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]:
         headers["HTTP-Referer"] = "https://openwebui.com/"
         headers["X-Title"] = "Open WebUI"
+    if ENABLE_FORWARD_USER_INFO_HEADERS:
+        headers["X-OpenWebUI-User-Name"] = user.name
+        headers["X-OpenWebUI-User-Id"] = user.id
+        headers["X-OpenWebUI-User-Email"] = user.email
+        headers["X-OpenWebUI-User-Role"] = user.role
 
     r = None
     session = None
@@ -505,6 +660,11 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
     headers = {}
     headers["Authorization"] = f"Bearer {key}"
     headers["Content-Type"] = "application/json"
+    if ENABLE_FORWARD_USER_INFO_HEADERS:
+        headers["X-OpenWebUI-User-Name"] = user.name
+        headers["X-OpenWebUI-User-Id"] = user.id
+        headers["X-OpenWebUI-User-Email"] = user.email
+        headers["X-OpenWebUI-User-Role"] = user.role
 
     r = None
     session = None

backend/open_webui/apps/retrieval/loaders/main.py

@@ -159,7 +159,7 @@ class Loader:
             elif file_ext in ["htm", "html"]:
                 loader = BSHTMLLoader(file_path, open_encoding="unicode_escape")
             elif file_ext == "md":
-                loader = UnstructuredMarkdownLoader(file_path)
+                loader = TextLoader(file_path, autodetect_encoding=True)
             elif file_content_type == "application/epub+zip":
                 loader = UnstructuredEPubLoader(file_path)
             elif (

backend/open_webui/apps/retrieval/loaders/youtube.py

@@ -0,0 +1,98 @@
+from typing import Any, Dict, Generator, List, Optional, Sequence, Union
+from urllib.parse import parse_qs, urlparse
+from langchain_core.documents import Document
+
+
+ALLOWED_SCHEMES = {"http", "https"}
+ALLOWED_NETLOCS = {
+    "youtu.be",
+    "m.youtube.com",
+    "youtube.com",
+    "www.youtube.com",
+    "www.youtube-nocookie.com",
+    "vid.plus",
+}
+
+
+def _parse_video_id(url: str) -> Optional[str]:
+    """Parse a YouTube URL and return the video ID if valid, otherwise None."""
+    parsed_url = urlparse(url)
+
+    if parsed_url.scheme not in ALLOWED_SCHEMES:
+        return None
+
+    if parsed_url.netloc not in ALLOWED_NETLOCS:
+        return None
+
+    path = parsed_url.path
+
+    if path.endswith("/watch"):
+        query = parsed_url.query
+        parsed_query = parse_qs(query)
+        if "v" in parsed_query:
+            ids = parsed_query["v"]
+            video_id = ids if isinstance(ids, str) else ids[0]
+        else:
+            return None
+    else:
+        path = parsed_url.path.lstrip("/")
+        video_id = path.split("/")[-1]
+
+    if len(video_id) != 11:  # Video IDs are 11 characters long
+        return None
+
+    return video_id
+
+
+class YoutubeLoader:
+    """Load `YouTube` video transcripts."""
+
+    def __init__(
+        self,
+        video_id: str,
+        language: Union[str, Sequence[str]] = "en",
+    ):
+        """Initialize with YouTube video ID."""
+        _video_id = _parse_video_id(video_id)
+        self.video_id = _video_id if _video_id is not None else video_id
+        self._metadata = {"source": video_id}
+        self.language = language
+        if isinstance(language, str):
+            self.language = [language]
+        else:
+            self.language = language
+
+    def load(self) -> List[Document]:
+        """Load YouTube transcripts into `Document` objects."""
+        try:
+            from youtube_transcript_api import (
+                NoTranscriptFound,
+                TranscriptsDisabled,
+                YouTubeTranscriptApi,
+            )
+        except ImportError:
+            raise ImportError(
+                'Could not import "youtube_transcript_api" Python package. '
+                "Please install it with `pip install youtube-transcript-api`."
+            )
+
+        try:
+            transcript_list = YouTubeTranscriptApi.list_transcripts(self.video_id)
+        except Exception as e:
+            print(e)
+            return []
+
+        try:
+            transcript = transcript_list.find_transcript(self.language)
+        except NoTranscriptFound:
+            transcript = transcript_list.find_transcript(["en"])
+
+        transcript_pieces: List[Dict[str, Any]] = transcript.fetch()
+
+        transcript = " ".join(
+            map(
+                lambda transcript_piece: transcript_piece["text"].strip(" "),
+                transcript_pieces,
+            )
+        )
+        return [Document(page_content=transcript, metadata=self._metadata)]

backend/open_webui/apps/retrieval/main.py

@@ -23,6 +23,7 @@ from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
 
 # Document loaders
 from open_webui.apps.retrieval.loaders.main import Loader
+from open_webui.apps.retrieval.loaders.youtube import YoutubeLoader
 
 # Web search engines
 from open_webui.apps.retrieval.web.main import SearchResult
@@ -38,6 +39,7 @@ from open_webui.apps.retrieval.web.serper import search_serper
 from open_webui.apps.retrieval.web.serply import search_serply
 from open_webui.apps.retrieval.web.serpstack import search_serpstack
 from open_webui.apps.retrieval.web.tavily import search_tavily
+from open_webui.apps.retrieval.web.bing import search_bing
 
 
 from open_webui.apps.retrieval.utils import (
@@ -76,6 +78,8 @@ from open_webui.config import (
     RAG_FILE_MAX_SIZE,
     RAG_OPENAI_API_BASE_URL,
     RAG_OPENAI_API_KEY,
+    RAG_OLLAMA_BASE_URL,
+    RAG_OLLAMA_API_KEY,
     RAG_RELEVANCE_THRESHOLD,
     RAG_RERANKING_MODEL,
     RAG_RERANKING_MODEL_AUTO_UPDATE,
@@ -87,6 +91,7 @@ from open_webui.config import (
     RAG_WEB_SEARCH_DOMAIN_FILTER_LIST,
     RAG_WEB_SEARCH_ENGINE,
     RAG_WEB_SEARCH_RESULT_COUNT,
+    JINA_API_KEY,
     SEARCHAPI_API_KEY,
     SEARCHAPI_ENGINE,
     SEARXNG_QUERY_URL,
@@ -95,13 +100,20 @@ from open_webui.config import (
     SERPSTACK_API_KEY,
     SERPSTACK_HTTPS,
     TAVILY_API_KEY,
+    BING_SEARCH_V7_ENDPOINT,
+    BING_SEARCH_V7_SUBSCRIPTION_KEY,
     TIKA_SERVER_URL,
     UPLOAD_DIR,
     YOUTUBE_LOADER_LANGUAGE,
+    DEFAULT_LOCALE,
     AppConfig,
 )
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.env import SRC_LOG_LEVELS, DEVICE_TYPE, DOCKER
+from open_webui.env import (
+    SRC_LOG_LEVELS,
+    DEVICE_TYPE,
+    DOCKER,
+)
 from open_webui.utils.misc import (
     calculate_sha256,
     calculate_sha256_string,
@@ -111,16 +123,17 @@ from open_webui.utils.misc import (
 from open_webui.utils.utils import get_admin_user, get_verified_user
 
 from langchain.text_splitter import RecursiveCharacterTextSplitter, TokenTextSplitter
-from langchain_community.document_loaders import (
-    YoutubeLoader,
-)
 from langchain_core.documents import Document
 
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
-app = FastAPI()
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
 
 app.state.config = AppConfig()
 
@@ -152,6 +165,9 @@ app.state.config.RAG_TEMPLATE = RAG_TEMPLATE
 app.state.config.OPENAI_API_BASE_URL = RAG_OPENAI_API_BASE_URL
 app.state.config.OPENAI_API_KEY = RAG_OPENAI_API_KEY
 
+app.state.config.OLLAMA_BASE_URL = RAG_OLLAMA_BASE_URL
+app.state.config.OLLAMA_API_KEY = RAG_OLLAMA_API_KEY
+
 app.state.config.PDF_EXTRACT_IMAGES = PDF_EXTRACT_IMAGES
 
 app.state.config.YOUTUBE_LOADER_LANGUAGE = YOUTUBE_LOADER_LANGUAGE
@@ -174,6 +190,10 @@ app.state.config.SERPLY_API_KEY = SERPLY_API_KEY
 app.state.config.TAVILY_API_KEY = TAVILY_API_KEY
 app.state.config.SEARCHAPI_API_KEY = SEARCHAPI_API_KEY
 app.state.config.SEARCHAPI_ENGINE = SEARCHAPI_ENGINE
+app.state.config.JINA_API_KEY = JINA_API_KEY
+app.state.config.BING_SEARCH_V7_ENDPOINT = BING_SEARCH_V7_ENDPOINT
+app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = BING_SEARCH_V7_SUBSCRIPTION_KEY
+
 app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = RAG_WEB_SEARCH_RESULT_COUNT
 app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = RAG_WEB_SEARCH_CONCURRENT_REQUESTS
 
@@ -185,11 +205,15 @@ def update_embedding_model(
     if embedding_model and app.state.config.RAG_EMBEDDING_ENGINE == "":
         from sentence_transformers import SentenceTransformer
 
-        app.state.sentence_transformer_ef = SentenceTransformer(
-            get_model_path(embedding_model, auto_update),
-            device=DEVICE_TYPE,
-            trust_remote_code=RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
-        )
+        try:
+            app.state.sentence_transformer_ef = SentenceTransformer(
+                get_model_path(embedding_model, auto_update),
+                device=DEVICE_TYPE,
+                trust_remote_code=RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
+            )
+        except Exception as e:
+            log.debug(f"Error loading SentenceTransformer: {e}")
+            app.state.sentence_transformer_ef = None
     else:
         app.state.sentence_transformer_ef = None
 
@@ -243,8 +267,16 @@ app.state.EMBEDDING_FUNCTION = get_embedding_function(
     app.state.config.RAG_EMBEDDING_ENGINE,
     app.state.config.RAG_EMBEDDING_MODEL,
     app.state.sentence_transformer_ef,
-    app.state.config.OPENAI_API_KEY,
-    app.state.config.OPENAI_API_BASE_URL,
+    (
+        app.state.config.OPENAI_API_BASE_URL
+        if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+        else app.state.config.OLLAMA_BASE_URL
+    ),
+    (
+        app.state.config.OPENAI_API_KEY
+        if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+        else app.state.config.OLLAMA_API_KEY
+    ),
     app.state.config.RAG_EMBEDDING_BATCH_SIZE,
 )
 
@@ -294,6 +326,10 @@ async def get_embedding_config(user=Depends(get_admin_user)):
             "url": app.state.config.OPENAI_API_BASE_URL,
             "key": app.state.config.OPENAI_API_KEY,
         },
+        "ollama_config": {
+            "url": app.state.config.OLLAMA_BASE_URL,
+            "key": app.state.config.OLLAMA_API_KEY,
+        },
     }
 
 
@@ -310,8 +346,14 @@ class OpenAIConfigForm(BaseModel):
     key: str
 
 
+class OllamaConfigForm(BaseModel):
+    url: str
+    key: str
+
+
 class EmbeddingModelUpdateForm(BaseModel):
     openai_config: Optional[OpenAIConfigForm] = None
+    ollama_config: Optional[OllamaConfigForm] = None
     embedding_engine: str
     embedding_model: str
     embedding_batch_size: Optional[int] = 1
@@ -332,6 +374,11 @@ async def update_embedding_config(
             if form_data.openai_config is not None:
                 app.state.config.OPENAI_API_BASE_URL = form_data.openai_config.url
                 app.state.config.OPENAI_API_KEY = form_data.openai_config.key
+
+            if form_data.ollama_config is not None:
+                app.state.config.OLLAMA_BASE_URL = form_data.ollama_config.url
+                app.state.config.OLLAMA_API_KEY = form_data.ollama_config.key
+
             app.state.config.RAG_EMBEDDING_BATCH_SIZE = form_data.embedding_batch_size
 
         update_embedding_model(app.state.config.RAG_EMBEDDING_MODEL)
@@ -340,8 +387,16 @@ async def update_embedding_config(
             app.state.config.RAG_EMBEDDING_ENGINE,
             app.state.config.RAG_EMBEDDING_MODEL,
             app.state.sentence_transformer_ef,
-            app.state.config.OPENAI_API_KEY,
-            app.state.config.OPENAI_API_BASE_URL,
+            (
+                app.state.config.OPENAI_API_BASE_URL
+                if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+                else app.state.config.OLLAMA_BASE_URL
+            ),
+            (
+                app.state.config.OPENAI_API_KEY
+                if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+                else app.state.config.OLLAMA_API_KEY
+            ),
             app.state.config.RAG_EMBEDDING_BATCH_SIZE,
         )
 
@@ -354,6 +409,10 @@ async def update_embedding_config(
                 "url": app.state.config.OPENAI_API_BASE_URL,
                 "key": app.state.config.OPENAI_API_KEY,
             },
+            "ollama_config": {
+                "url": app.state.config.OLLAMA_BASE_URL,
+                "key": app.state.config.OLLAMA_API_KEY,
+            },
         }
     except Exception as e:
         log.exception(f"Problem updating embedding model: {e}")
@@ -414,7 +473,7 @@ async def get_rag_config(user=Depends(get_admin_user)):
             "translation": app.state.YOUTUBE_LOADER_TRANSLATION,
         },
         "web": {
-            "ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+            "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
             "search": {
                 "enabled": app.state.config.ENABLE_RAG_WEB_SEARCH,
                 "engine": app.state.config.RAG_WEB_SEARCH_ENGINE,
@@ -430,6 +489,9 @@ async def get_rag_config(user=Depends(get_admin_user)):
                 "tavily_api_key": app.state.config.TAVILY_API_KEY,
                 "searchapi_api_key": app.state.config.SEARCHAPI_API_KEY,
                 "seaarchapi_engine": app.state.config.SEARCHAPI_ENGINE,
+                "jina_api_key": app.state.config.JINA_API_KEY,
+                "bing_search_v7_endpoint": app.state.config.BING_SEARCH_V7_ENDPOINT,
+                "bing_search_v7_subscription_key": app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY,
                 "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
                 "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
             },
@@ -473,6 +535,9 @@ class WebSearchConfig(BaseModel):
     tavily_api_key: Optional[str] = None
     searchapi_api_key: Optional[str] = None
     searchapi_engine: Optional[str] = None
+    jina_api_key: Optional[str] = None
+    bing_search_v7_endpoint: Optional[str] = None
+    bing_search_v7_subscription_key: Optional[str] = None
     result_count: Optional[int] = None
     concurrent_requests: Optional[int] = None
 
@@ -519,6 +584,7 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
 
     if form_data.web is not None:
         app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = (
+            # Note: When UI "Bypass SSL verification for Websites"=True then ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION=False
             form_data.web.web_loader_ssl_verification
         )
 
@@ -540,6 +606,15 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
         app.state.config.TAVILY_API_KEY = form_data.web.search.tavily_api_key
         app.state.config.SEARCHAPI_API_KEY = form_data.web.search.searchapi_api_key
         app.state.config.SEARCHAPI_ENGINE = form_data.web.search.searchapi_engine
+
+        app.state.config.JINA_API_KEY = form_data.web.search.jina_api_key
+        app.state.config.BING_SEARCH_V7_ENDPOINT = (
+            form_data.web.search.bing_search_v7_endpoint
+        )
+        app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = (
+            form_data.web.search.bing_search_v7_subscription_key
+        )
+
         app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = form_data.web.search.result_count
         app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = (
             form_data.web.search.concurrent_requests
@@ -566,7 +641,7 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
             "translation": app.state.YOUTUBE_LOADER_TRANSLATION,
         },
         "web": {
-            "ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+            "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
             "search": {
                 "enabled": app.state.config.ENABLE_RAG_WEB_SEARCH,
                 "engine": app.state.config.RAG_WEB_SEARCH_ENGINE,
@@ -582,6 +657,9 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
                 "serachapi_api_key": app.state.config.SEARCHAPI_API_KEY,
                 "searchapi_engine": app.state.config.SEARCHAPI_ENGINE,
                 "tavily_api_key": app.state.config.TAVILY_API_KEY,
+                "jina_api_key": app.state.config.JINA_API_KEY,
+                "bing_search_v7_endpoint": app.state.config.BING_SEARCH_V7_ENDPOINT,
+                "bing_search_v7_subscription_key": app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY,
                 "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
                 "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
             },
@@ -643,6 +721,23 @@ async def update_query_settings(
 ####################################
 
 
+def _get_docs_info(docs: list[Document]) -> str:
+    docs_info = set()
+
+    # Trying to select relevant metadata identifying the document.
+    for doc in docs:
+        metadata = getattr(doc, "metadata", {})
+        doc_name = metadata.get("name", "")
+        if not doc_name:
+            doc_name = metadata.get("title", "")
+        if not doc_name:
+            doc_name = metadata.get("source", "")
+        if doc_name:
+            docs_info.add(doc_name)
+
+    return ", ".join(docs_info)
+
+
 def save_docs_to_vector_db(
     docs,
     collection_name,
@@ -651,7 +746,9 @@ def save_docs_to_vector_db(
     split: bool = True,
     add: bool = False,
 ) -> bool:
-    log.info(f"save_docs_to_vector_db {docs} {collection_name}")
+    log.info(
+        f"save_docs_to_vector_db: document {_get_docs_info(docs)} {collection_name}"
+    )
 
     # Check if entries with the same hash (metadata.hash) already exist
     if metadata and "hash" in metadata:
@@ -733,8 +830,16 @@ def save_docs_to_vector_db(
             app.state.config.RAG_EMBEDDING_ENGINE,
             app.state.config.RAG_EMBEDDING_MODEL,
             app.state.sentence_transformer_ef,
-            app.state.config.OPENAI_API_KEY,
-            app.state.config.OPENAI_API_BASE_URL,
+            (
+                app.state.config.OPENAI_API_BASE_URL
+                if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+                else app.state.config.OLLAMA_BASE_URL
+            ),
+            (
+                app.state.config.OPENAI_API_KEY
+                if app.state.config.RAG_EMBEDDING_ENGINE == "openai"
+                else app.state.config.OLLAMA_API_KEY
+            ),
             app.state.config.RAG_EMBEDDING_BATCH_SIZE,
         )
 
@@ -959,12 +1064,10 @@ def process_youtube_video(form_data: ProcessUrlForm, user=Depends(get_verified_u
         if not collection_name:
             collection_name = calculate_sha256_string(form_data.url)[:63]
 
-        loader = YoutubeLoader.from_youtube_url(
-            form_data.url,
-            add_video_info=True,
-            language=app.state.config.YOUTUBE_LOADER_LANGUAGE,
-            translation=app.state.YOUTUBE_LOADER_TRANSLATION,
+        loader = YoutubeLoader(
+            form_data.url, language=app.state.config.YOUTUBE_LOADER_LANGUAGE
         )
+
         docs = loader.load()
         content = " ".join([doc.page_content for doc in docs])
         log.debug(f"text_content: {content}")
@@ -1150,7 +1253,20 @@ def search_web(engine: str, query: str) -> list[SearchResult]:
         else:
             raise Exception("No SEARCHAPI_API_KEY found in environment variables")
     elif engine == "jina":
-        return search_jina(query, app.state.config.RAG_WEB_SEARCH_RESULT_COUNT)
+        return search_jina(
+            app.state.config.JINA_API_KEY,
+            query,
+            app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
+        )
+    elif engine == "bing":
+        return search_bing(
+            app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY,
+            app.state.config.BING_SEARCH_V7_ENDPOINT,
+            str(DEFAULT_LOCALE),
+            query,
+            app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
+            app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST,
+        )
     else:
         raise Exception("No search engine API key found in environment variables")
 
@@ -1180,8 +1296,12 @@ def process_web_search(form_data: SearchForm, user=Depends(get_verified_user)):
 
         urls = [result.link for result in web_results]
 
-        loader = get_web_loader(urls)
-        docs = loader.load()
+        loader = get_web_loader(
+            urls,
+            verify_ssl=app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+            requests_per_second=app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
+        )
+        docs = loader.aload()
 
         save_docs_to_vector_db(docs, collection_name, overwrite=True)
 

backend/open_webui/apps/retrieval/utils.py

@@ -3,6 +3,7 @@ import os
 import uuid
 from typing import Optional, Union
 
+import asyncio
 import requests
 
 from huggingface_hub import snapshot_download
@@ -10,11 +11,6 @@ from langchain.retrievers import ContextualCompressionRetriever, EnsembleRetriev
 from langchain_community.retrievers import BM25Retriever
 from langchain_core.documents import Document
 
-
-from open_webui.apps.ollama.main import (
-    GenerateEmbedForm,
-    generate_ollama_batch_embeddings,
-)
 from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
 from open_webui.utils.misc import get_last_user_message
 
@@ -76,7 +72,7 @@ def query_doc(
             limit=k,
         )
 
-        log.info(f"query_doc:result {result}")
+        log.info(f"query_doc:result {result.ids} {result.metadatas}")
         return result
     except Exception as e:
         print(e)
@@ -127,7 +123,10 @@ def query_doc_with_hybrid_search(
             "metadatas": [[d.metadata for d in result]],
         }
 
-        log.info(f"query_doc_with_hybrid_search:result {result}")
+        log.info(
+            "query_doc_with_hybrid_search:result "
+            + f'{result["metadatas"]} {result["distances"]}'
+        )
         return result
     except Exception as e:
         raise e
@@ -178,35 +177,34 @@ def merge_and_sort_query_results(
 
 def query_collection(
     collection_names: list[str],
-    query: str,
+    queries: list[str],
     embedding_function,
     k: int,
 ) -> dict:
-
     results = []
-    query_embedding = embedding_function(query)
-
-    for collection_name in collection_names:
-        if collection_name:
-            try:
-                result = query_doc(
-                    collection_name=collection_name,
-                    k=k,
-                    query_embedding=query_embedding,
-                )
-                if result is not None:
-                    results.append(result.model_dump())
-            except Exception as e:
-                log.exception(f"Error when querying the collection: {e}")
-        else:
-            pass
+    for query in queries:
+        query_embedding = embedding_function(query)
+        for collection_name in collection_names:
+            if collection_name:
+                try:
+                    result = query_doc(
+                        collection_name=collection_name,
+                        k=k,
+                        query_embedding=query_embedding,
+                    )
+                    if result is not None:
+                        results.append(result.model_dump())
+                except Exception as e:
+                    log.exception(f"Error when querying the collection: {e}")
+            else:
+                pass
 
     return merge_and_sort_query_results(results, k=k)
 
 
 def query_collection_with_hybrid_search(
     collection_names: list[str],
-    query: str,
+    queries: list[str],
     embedding_function,
     k: int,
     reranking_function,
@@ -216,15 +214,16 @@ def query_collection_with_hybrid_search(
     error = False
     for collection_name in collection_names:
         try:
-            result = query_doc_with_hybrid_search(
-                collection_name=collection_name,
-                query=query,
-                embedding_function=embedding_function,
-                k=k,
-                reranking_function=reranking_function,
-                r=r,
-            )
-            results.append(result)
+            for query in queries:
+                result = query_doc_with_hybrid_search(
+                    collection_name=collection_name,
+                    query=query,
+                    embedding_function=embedding_function,
+                    k=k,
+                    reranking_function=reranking_function,
+                    r=r,
+                )
+                results.append(result)
         except Exception as e:
             log.exception(
                 "Error when querying the collection with " f"hybrid_search: {e}"
@@ -281,8 +280,8 @@ def get_embedding_function(
     embedding_engine,
     embedding_model,
     embedding_function,
-    openai_key,
-    openai_url,
+    url,
+    key,
     embedding_batch_size,
 ):
     if embedding_engine == "":
@@ -292,8 +291,8 @@ def get_embedding_function(
             engine=embedding_engine,
             model=embedding_model,
             text=query,
-            key=openai_key if embedding_engine == "openai" else "",
-            url=openai_url if embedding_engine == "openai" else "",
+            url=url,
+            key=key,
         )
 
         def generate_multiple(query, func):
@@ -310,15 +309,14 @@ def get_embedding_function(
 
 def get_rag_context(
     files,
-    messages,
+    queries,
     embedding_function,
     k,
     reranking_function,
     r,
     hybrid_search,
 ):
-    log.debug(f"files: {files} {messages} {embedding_function} {reranking_function}")
-    query = get_last_user_message(messages)
+    log.debug(f"files: {files} {queries} {embedding_function} {reranking_function}")
 
     extracted_collections = []
     relevant_contexts = []
@@ -360,7 +358,7 @@ def get_rag_context(
                         try:
                             context = query_collection_with_hybrid_search(
                                 collection_names=collection_names,
-                                query=query,
+                                queries=queries,
                                 embedding_function=embedding_function,
                                 k=k,
                                 reranking_function=reranking_function,
@@ -375,7 +373,7 @@ def get_rag_context(
                     if (not hybrid_search) or (context is None):
                         context = query_collection(
                             collection_names=collection_names,
-                            query=query,
+                            queries=queries,
                             embedding_function=embedding_function,
                             k=k,
                         )
@@ -467,7 +465,7 @@ def get_model_path(model: str, update_model: bool = False):
 
 
 def generate_openai_batch_embeddings(
-    model: str, texts: list[str], key: str, url: str = "https://api.openai.com/v1"
+    model: str, texts: list[str], url: str = "https://api.openai.com/v1", key: str = ""
 ) -> Optional[list[list[float]]]:
     try:
         r = requests.post(
@@ -489,29 +487,50 @@ def generate_openai_batch_embeddings(
         return None
 
 
+def generate_ollama_batch_embeddings(
+    model: str, texts: list[str], url: str, key: str
+) -> Optional[list[list[float]]]:
+    try:
+        r = requests.post(
+            f"{url}/api/embed",
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {key}",
+            },
+            json={"input": texts, "model": model},
+        )
+        r.raise_for_status()
+        data = r.json()
+
+        print(data)
+        if "embeddings" in data:
+            return data["embeddings"]
+        else:
+            raise "Something went wrong :/"
+    except Exception as e:
+        print(e)
+        return None
+
+
 def generate_embeddings(engine: str, model: str, text: Union[str, list[str]], **kwargs):
+    url = kwargs.get("url", "")
+    key = kwargs.get("key", "")
+
     if engine == "ollama":
         if isinstance(text, list):
             embeddings = generate_ollama_batch_embeddings(
-                GenerateEmbedForm(**{"model": model, "input": text})
+                **{"model": model, "texts": text, "url": url, "key": key}
             )
         else:
             embeddings = generate_ollama_batch_embeddings(
-                GenerateEmbedForm(**{"model": model, "input": [text]})
+                **{"model": model, "texts": [text], "url": url, "key": key}
             )
-        return (
-            embeddings["embeddings"][0]
-            if isinstance(text, str)
-            else embeddings["embeddings"]
-        )
+        return embeddings[0] if isinstance(text, str) else embeddings
     elif engine == "openai":
-        key = kwargs.get("key", "")
-        url = kwargs.get("url", "https://api.openai.com/v1")
-
         if isinstance(text, list):
-            embeddings = generate_openai_batch_embeddings(model, text, key, url)
+            embeddings = generate_openai_batch_embeddings(model, text, url, key)
         else:
-            embeddings = generate_openai_batch_embeddings(model, [text], key, url)
+            embeddings = generate_openai_batch_embeddings(model, [text], url, key)
 
         return embeddings[0] if isinstance(text, str) else embeddings
 

backend/open_webui/apps/retrieval/vector/connector.py

@@ -8,6 +8,14 @@ elif VECTOR_DB == "qdrant":
     from open_webui.apps.retrieval.vector.dbs.qdrant import QdrantClient
 
     VECTOR_DB_CLIENT = QdrantClient()
+elif VECTOR_DB == "opensearch":
+    from open_webui.apps.retrieval.vector.dbs.opensearch import OpenSearchClient
+
+    VECTOR_DB_CLIENT = OpenSearchClient()
+elif VECTOR_DB == "pgvector":
+    from open_webui.apps.retrieval.vector.dbs.pgvector import PgvectorClient
+
+    VECTOR_DB_CLIENT = PgvectorClient()
 else:
     from open_webui.apps.retrieval.vector.dbs.chroma import ChromaClient
 

backend/open_webui/apps/retrieval/vector/dbs/chroma.py

@@ -13,11 +13,24 @@ from open_webui.config import (
     CHROMA_HTTP_SSL,
     CHROMA_TENANT,
     CHROMA_DATABASE,
+    CHROMA_CLIENT_AUTH_PROVIDER,
+    CHROMA_CLIENT_AUTH_CREDENTIALS,
 )
 
 
 class ChromaClient:
     def __init__(self):
+        settings_dict = {
+            "allow_reset": True,
+            "anonymized_telemetry": False,
+        }
+        if CHROMA_CLIENT_AUTH_PROVIDER is not None:
+            settings_dict["chroma_client_auth_provider"] = CHROMA_CLIENT_AUTH_PROVIDER
+        if CHROMA_CLIENT_AUTH_CREDENTIALS is not None:
+            settings_dict["chroma_client_auth_credentials"] = (
+                CHROMA_CLIENT_AUTH_CREDENTIALS
+            )
+
         if CHROMA_HTTP_HOST != "":
             self.client = chromadb.HttpClient(
                 host=CHROMA_HTTP_HOST,
@@ -26,12 +39,12 @@ class ChromaClient:
                 ssl=CHROMA_HTTP_SSL,
                 tenant=CHROMA_TENANT,
                 database=CHROMA_DATABASE,
-                settings=Settings(allow_reset=True, anonymized_telemetry=False),
+                settings=Settings(**settings_dict),
             )
         else:
             self.client = chromadb.PersistentClient(
                 path=CHROMA_DATA_PATH,
-                settings=Settings(allow_reset=True, anonymized_telemetry=False),
+                settings=Settings(**settings_dict),
                 tenant=CHROMA_TENANT,
                 database=CHROMA_DATABASE,
             )

backend/open_webui/apps/retrieval/vector/dbs/opensearch.py

@@ -0,0 +1,178 @@
+from opensearchpy import OpenSearch
+from typing import Optional
+
+from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.config import (
+    OPENSEARCH_URI,
+    OPENSEARCH_SSL,
+    OPENSEARCH_CERT_VERIFY,
+    OPENSEARCH_USERNAME,
+    OPENSEARCH_PASSWORD,
+)
+
+
+class OpenSearchClient:
+    def __init__(self):
+        self.index_prefix = "open_webui"
+        self.client = OpenSearch(
+            hosts=[OPENSEARCH_URI],
+            use_ssl=OPENSEARCH_SSL,
+            verify_certs=OPENSEARCH_CERT_VERIFY,
+            http_auth=(OPENSEARCH_USERNAME, OPENSEARCH_PASSWORD),
+        )
+
+    def _result_to_get_result(self, result) -> GetResult:
+        ids = []
+        documents = []
+        metadatas = []
+
+        for hit in result["hits"]["hits"]:
+            ids.append(hit["_id"])
+            documents.append(hit["_source"].get("text"))
+            metadatas.append(hit["_source"].get("metadata"))
+
+        return GetResult(ids=ids, documents=documents, metadatas=metadatas)
+
+    def _result_to_search_result(self, result) -> SearchResult:
+        ids = []
+        distances = []
+        documents = []
+        metadatas = []
+
+        for hit in result["hits"]["hits"]:
+            ids.append(hit["_id"])
+            distances.append(hit["_score"])
+            documents.append(hit["_source"].get("text"))
+            metadatas.append(hit["_source"].get("metadata"))
+
+        return SearchResult(
+            ids=ids, distances=distances, documents=documents, metadatas=metadatas
+        )
+
+    def _create_index(self, index_name: str, dimension: int):
+        body = {
+            "mappings": {
+                "properties": {
+                    "id": {"type": "keyword"},
+                    "vector": {
+                        "type": "dense_vector",
+                        "dims": dimension,  # Adjust based on your vector dimensions
+                        "index": true,
+                        "similarity": "faiss",
+                        "method": {
+                            "name": "hnsw",
+                            "space_type": "ip",  # Use inner product to approximate cosine similarity
+                            "engine": "faiss",
+                            "ef_construction": 128,
+                            "m": 16,
+                        },
+                    },
+                    "text": {"type": "text"},
+                    "metadata": {"type": "object"},
+                }
+            }
+        }
+        self.client.indices.create(index=f"{self.index_prefix}_{index_name}", body=body)
+
+    def _create_batches(self, items: list[VectorItem], batch_size=100):
+        for i in range(0, len(items), batch_size):
+            yield items[i : i + batch_size]
+
+    def has_collection(self, index_name: str) -> bool:
+        # has_collection here means has index.
+        # We are simply adapting to the norms of the other DBs.
+        return self.client.indices.exists(index=f"{self.index_prefix}_{index_name}")
+
+    def delete_colleciton(self, index_name: str):
+        # delete_collection here means delete index.
+        # We are simply adapting to the norms of the other DBs.
+        self.client.indices.delete(index=f"{self.index_prefix}_{index_name}")
+
+    def search(
+        self, index_name: str, vectors: list[list[float]], limit: int
+    ) -> Optional[SearchResult]:
+        query = {
+            "size": limit,
+            "_source": ["text", "metadata"],
+            "query": {
+                "script_score": {
+                    "query": {"match_all": {}},
+                    "script": {
+                        "source": "cosineSimilarity(params.vector, 'vector') + 1.0",
+                        "params": {
+                            "vector": vectors[0]
+                        },  # Assuming single query vector
+                    },
+                }
+            },
+        }
+
+        result = self.client.search(
+            index=f"{self.index_prefix}_{index_name}", body=query
+        )
+
+        return self._result_to_search_result(result)
+
+    def get_or_create_index(self, index_name: str, dimension: int):
+        if not self.has_index(index_name):
+            self._create_index(index_name, dimension)
+
+    def get(self, index_name: str) -> Optional[GetResult]:
+        query = {"query": {"match_all": {}}, "_source": ["text", "metadata"]}
+
+        result = self.client.search(
+            index=f"{self.index_prefix}_{index_name}", body=query
+        )
+        return self._result_to_get_result(result)
+
+    def insert(self, index_name: str, items: list[VectorItem]):
+        if not self.has_index(index_name):
+            self._create_index(index_name, dimension=len(items[0]["vector"]))
+
+        for batch in self._create_batches(items):
+            actions = [
+                {
+                    "index": {
+                        "_id": item["id"],
+                        "_source": {
+                            "vector": item["vector"],
+                            "text": item["text"],
+                            "metadata": item["metadata"],
+                        },
+                    }
+                }
+                for item in batch
+            ]
+            self.client.bulk(actions)
+
+    def upsert(self, index_name: str, items: list[VectorItem]):
+        if not self.has_index(index_name):
+            self._create_index(index_name, dimension=len(items[0]["vector"]))
+
+        for batch in self._create_batches(items):
+            actions = [
+                {
+                    "index": {
+                        "_id": item["id"],
+                        "_source": {
+                            "vector": item["vector"],
+                            "text": item["text"],
+                            "metadata": item["metadata"],
+                        },
+                    }
+                }
+                for item in batch
+            ]
+            self.client.bulk(actions)
+
+    def delete(self, index_name: str, ids: list[str]):
+        actions = [
+            {"delete": {"_index": f"{self.index_prefix}_{index_name}", "_id": id}}
+            for id in ids
+        ]
+        self.client.bulk(body=actions)
+
+    def reset(self):
+        indices = self.client.indices.get(index=f"{self.index_prefix}_*")
+        for index in indices:
+            self.client.indices.delete(index=index)

backend/open_webui/apps/retrieval/vector/dbs/pgvector.py

@@ -0,0 +1,354 @@
+from typing import Optional, List, Dict, Any
+from sqlalchemy import (
+    cast,
+    column,
+    create_engine,
+    Column,
+    Integer,
+    select,
+    text,
+    Text,
+    values,
+)
+from sqlalchemy.sql import true
+from sqlalchemy.pool import NullPool
+
+from sqlalchemy.orm import declarative_base, scoped_session, sessionmaker
+from sqlalchemy.dialects.postgresql import JSONB, array
+from pgvector.sqlalchemy import Vector
+from sqlalchemy.ext.mutable import MutableDict
+
+from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.config import PGVECTOR_DB_URL
+
+VECTOR_LENGTH = 1536
+Base = declarative_base()
+
+
+class DocumentChunk(Base):
+    __tablename__ = "document_chunk"
+
+    id = Column(Text, primary_key=True)
+    vector = Column(Vector(dim=VECTOR_LENGTH), nullable=True)
+    collection_name = Column(Text, nullable=False)
+    text = Column(Text, nullable=True)
+    vmetadata = Column(MutableDict.as_mutable(JSONB), nullable=True)
+
+
+class PgvectorClient:
+    def __init__(self) -> None:
+
+        # if no pgvector uri, use the existing database connection
+        if not PGVECTOR_DB_URL:
+            from open_webui.apps.webui.internal.db import Session
+
+            self.session = Session
+        else:
+            engine = create_engine(
+                PGVECTOR_DB_URL, pool_pre_ping=True, poolclass=NullPool
+            )
+            SessionLocal = sessionmaker(
+                autocommit=False, autoflush=False, bind=engine, expire_on_commit=False
+            )
+            self.session = scoped_session(SessionLocal)
+
+        try:
+            # Ensure the pgvector extension is available
+            self.session.execute(text("CREATE EXTENSION IF NOT EXISTS vector;"))
+
+            # Create the tables if they do not exist
+            # Base.metadata.create_all requires a bind (engine or connection)
+            # Get the connection from the session
+            connection = self.session.connection()
+            Base.metadata.create_all(bind=connection)
+
+            # Create an index on the vector column if it doesn't exist
+            self.session.execute(
+                text(
+                    "CREATE INDEX IF NOT EXISTS idx_document_chunk_vector "
+                    "ON document_chunk USING ivfflat (vector vector_cosine_ops) WITH (lists = 100);"
+                )
+            )
+            self.session.execute(
+                text(
+                    "CREATE INDEX IF NOT EXISTS idx_document_chunk_collection_name "
+                    "ON document_chunk (collection_name);"
+                )
+            )
+            self.session.commit()
+            print("Initialization complete.")
+        except Exception as e:
+            self.session.rollback()
+            print(f"Error during initialization: {e}")
+            raise
+
+    def adjust_vector_length(self, vector: List[float]) -> List[float]:
+        # Adjust vector to have length VECTOR_LENGTH
+        current_length = len(vector)
+        if current_length < VECTOR_LENGTH:
+            # Pad the vector with zeros
+            vector += [0.0] * (VECTOR_LENGTH - current_length)
+        elif current_length > VECTOR_LENGTH:
+            raise Exception(
+                f"Vector length {current_length} not supported. Max length must be <= {VECTOR_LENGTH}"
+            )
+        return vector
+
+    def insert(self, collection_name: str, items: List[VectorItem]) -> None:
+        try:
+            new_items = []
+            for item in items:
+                vector = self.adjust_vector_length(item["vector"])
+                new_chunk = DocumentChunk(
+                    id=item["id"],
+                    vector=vector,
+                    collection_name=collection_name,
+                    text=item["text"],
+                    vmetadata=item["metadata"],
+                )
+                new_items.append(new_chunk)
+            self.session.bulk_save_objects(new_items)
+            self.session.commit()
+            print(
+                f"Inserted {len(new_items)} items into collection '{collection_name}'."
+            )
+        except Exception as e:
+            self.session.rollback()
+            print(f"Error during insert: {e}")
+            raise
+
+    def upsert(self, collection_name: str, items: List[VectorItem]) -> None:
+        try:
+            for item in items:
+                vector = self.adjust_vector_length(item["vector"])
+                existing = (
+                    self.session.query(DocumentChunk)
+                    .filter(DocumentChunk.id == item["id"])
+                    .first()
+                )
+                if existing:
+                    existing.vector = vector
+                    existing.text = item["text"]
+                    existing.vmetadata = item["metadata"]
+                    existing.collection_name = (
+                        collection_name  # Update collection_name if necessary
+                    )
+                else:
+                    new_chunk = DocumentChunk(
+                        id=item["id"],
+                        vector=vector,
+                        collection_name=collection_name,
+                        text=item["text"],
+                        vmetadata=item["metadata"],
+                    )
+                    self.session.add(new_chunk)
+            self.session.commit()
+            print(f"Upserted {len(items)} items into collection '{collection_name}'.")
+        except Exception as e:
+            self.session.rollback()
+            print(f"Error during upsert: {e}")
+            raise
+
+    def search(
+        self,
+        collection_name: str,
+        vectors: List[List[float]],
+        limit: Optional[int] = None,
+    ) -> Optional[SearchResult]:
+        try:
+            if not vectors:
+                return None
+
+            # Adjust query vectors to VECTOR_LENGTH
+            vectors = [self.adjust_vector_length(vector) for vector in vectors]
+            num_queries = len(vectors)
+
+            def vector_expr(vector):
+                return cast(array(vector), Vector(VECTOR_LENGTH))
+
+            # Create the values for query vectors
+            qid_col = column("qid", Integer)
+            q_vector_col = column("q_vector", Vector(VECTOR_LENGTH))
+            query_vectors = (
+                values(qid_col, q_vector_col)
+                .data(
+                    [(idx, vector_expr(vector)) for idx, vector in enumerate(vectors)]
+                )
+                .alias("query_vectors")
+            )
+
+            # Build the lateral subquery for each query vector
+            subq = (
+                select(
+                    DocumentChunk.id,
+                    DocumentChunk.text,
+                    DocumentChunk.vmetadata,
+                    (
+                        DocumentChunk.vector.cosine_distance(query_vectors.c.q_vector)
+                    ).label("distance"),
+                )
+                .where(DocumentChunk.collection_name == collection_name)
+                .order_by(
+                    (DocumentChunk.vector.cosine_distance(query_vectors.c.q_vector))
+                )
+            )
+            if limit is not None:
+                subq = subq.limit(limit)
+            subq = subq.lateral("result")
+
+            # Build the main query by joining query_vectors and the lateral subquery
+            stmt = (
+                select(
+                    query_vectors.c.qid,
+                    subq.c.id,
+                    subq.c.text,
+                    subq.c.vmetadata,
+                    subq.c.distance,
+                )
+                .select_from(query_vectors)
+                .join(subq, true())
+                .order_by(query_vectors.c.qid, subq.c.distance)
+            )
+
+            result_proxy = self.session.execute(stmt)
+            results = result_proxy.all()
+
+            ids = [[] for _ in range(num_queries)]
+            distances = [[] for _ in range(num_queries)]
+            documents = [[] for _ in range(num_queries)]
+            metadatas = [[] for _ in range(num_queries)]
+
+            if not results:
+                return SearchResult(
+                    ids=ids,
+                    distances=distances,
+                    documents=documents,
+                    metadatas=metadatas,
+                )
+
+            for row in results:
+                qid = int(row.qid)
+                ids[qid].append(row.id)
+                distances[qid].append(row.distance)
+                documents[qid].append(row.text)
+                metadatas[qid].append(row.vmetadata)
+
+            return SearchResult(
+                ids=ids, distances=distances, documents=documents, metadatas=metadatas
+            )
+        except Exception as e:
+            print(f"Error during search: {e}")
+            return None
+
+    def query(
+        self, collection_name: str, filter: Dict[str, Any], limit: Optional[int] = None
+    ) -> Optional[GetResult]:
+        try:
+            query = self.session.query(DocumentChunk).filter(
+                DocumentChunk.collection_name == collection_name
+            )
+
+            for key, value in filter.items():
+                query = query.filter(DocumentChunk.vmetadata[key].astext == str(value))
+
+            if limit is not None:
+                query = query.limit(limit)
+
+            results = query.all()
+
+            if not results:
+                return None
+
+            ids = [[result.id for result in results]]
+            documents = [[result.text for result in results]]
+            metadatas = [[result.vmetadata for result in results]]
+
+            return GetResult(
+                ids=ids,
+                documents=documents,
+                metadatas=metadatas,
+            )
+        except Exception as e:
+            print(f"Error during query: {e}")
+            return None
+
+    def get(
+        self, collection_name: str, limit: Optional[int] = None
+    ) -> Optional[GetResult]:
+        try:
+            query = self.session.query(DocumentChunk).filter(
+                DocumentChunk.collection_name == collection_name
+            )
+            if limit is not None:
+                query = query.limit(limit)
+
+            results = query.all()
+
+            if not results:
+                return None
+
+            ids = [[result.id for result in results]]
+            documents = [[result.text for result in results]]
+            metadatas = [[result.vmetadata for result in results]]
+
+            return GetResult(ids=ids, documents=documents, metadatas=metadatas)
+        except Exception as e:
+            print(f"Error during get: {e}")
+            return None
+
+    def delete(
+        self,
+        collection_name: str,
+        ids: Optional[List[str]] = None,
+        filter: Optional[Dict[str, Any]] = None,
+    ) -> None:
+        try:
+            query = self.session.query(DocumentChunk).filter(
+                DocumentChunk.collection_name == collection_name
+            )
+            if ids:
+                query = query.filter(DocumentChunk.id.in_(ids))
+            if filter:
+                for key, value in filter.items():
+                    query = query.filter(
+                        DocumentChunk.vmetadata[key].astext == str(value)
+                    )
+            deleted = query.delete(synchronize_session=False)
+            self.session.commit()
+            print(f"Deleted {deleted} items from collection '{collection_name}'.")
+        except Exception as e:
+            self.session.rollback()
+            print(f"Error during delete: {e}")
+            raise
+
+    def reset(self) -> None:
+        try:
+            deleted = self.session.query(DocumentChunk).delete()
+            self.session.commit()
+            print(
+                f"Reset complete. Deleted {deleted} items from 'document_chunk' table."
+            )
+        except Exception as e:
+            self.session.rollback()
+            print(f"Error during reset: {e}")
+            raise
+
+    def close(self) -> None:
+        pass
+
+    def has_collection(self, collection_name: str) -> bool:
+        try:
+            exists = (
+                self.session.query(DocumentChunk)
+                .filter(DocumentChunk.collection_name == collection_name)
+                .first()
+                is not None
+            )
+            return exists
+        except Exception as e:
+            print(f"Error checking collection existence: {e}")
+            return False
+
+    def delete_collection(self, collection_name: str) -> None:
+        self.delete(collection_name)
+        print(f"Collection '{collection_name}' deleted.")

backend/open_webui/apps/retrieval/vector/dbs/qdrant.py

@@ -5,7 +5,7 @@ from qdrant_client.http.models import PointStruct
 from qdrant_client.models import models
 
 from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
-from open_webui.config import QDRANT_URI
+from open_webui.config import QDRANT_URI, QDRANT_API_KEY
 
 NO_LIMIT = 999999999
 
@@ -14,7 +14,12 @@ class QdrantClient:
     def __init__(self):
         self.collection_prefix = "open-webui"
         self.QDRANT_URI = QDRANT_URI
-        self.client = Qclient(url=self.QDRANT_URI) if self.QDRANT_URI else None
+        self.QDRANT_API_KEY = QDRANT_API_KEY
+        self.client = (
+            Qclient(url=self.QDRANT_URI, api_key=self.QDRANT_API_KEY)
+            if self.QDRANT_URI
+            else None
+        )
 
     def _result_to_get_result(self, points) -> GetResult:
         ids = []

backend/open_webui/apps/retrieval/web/bing.py

@@ -0,0 +1,73 @@
+import logging
+import os
+from pprint import pprint
+from typing import Optional
+import requests
+from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
+import argparse
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["RAG"])
+"""
+Documentation: https://docs.microsoft.com/en-us/bing/search-apis/bing-web-search/overview
+"""
+
+
+def search_bing(
+    subscription_key: str,
+    endpoint: str,
+    locale: str,
+    query: str,
+    count: int,
+    filter_list: Optional[list[str]] = None,
+) -> list[SearchResult]:
+    mkt = locale
+    params = {"q": query, "mkt": mkt, "answerCount": count}
+    headers = {"Ocp-Apim-Subscription-Key": subscription_key}
+
+    try:
+        response = requests.get(endpoint, headers=headers, params=params)
+        response.raise_for_status()
+        json_response = response.json()
+        results = json_response.get("webPages", {}).get("value", [])
+        if filter_list:
+            results = get_filtered_results(results, filter_list)
+        return [
+            SearchResult(
+                link=result["url"],
+                title=result.get("name"),
+                snippet=result.get("snippet"),
+            )
+            for result in results
+        ]
+    except Exception as ex:
+        log.error(f"Error: {ex}")
+        raise ex
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Search Bing from the command line.")
+    parser.add_argument(
+        "query",
+        type=str,
+        default="Top 10 international news today",
+        help="The search query.",
+    )
+    parser.add_argument(
+        "--count", type=int, default=10, help="Number of search results to return."
+    )
+    parser.add_argument(
+        "--filter", nargs="*", help="List of filters to apply to the search results."
+    )
+    parser.add_argument(
+        "--locale",
+        type=str,
+        default="en-US",
+        help="The locale to use for the search, maps to market in api",
+    )
+
+    args = parser.parse_args()
+
+    results = search_bing(args.locale, args.query, args.count, args.filter)
+    pprint(results)

backend/open_webui/apps/retrieval/web/jina_search.py

@@ -9,7 +9,7 @@ log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
-def search_jina(query: str, count: int) -> list[SearchResult]:
+def search_jina(api_key: str, query: str, count: int) -> list[SearchResult]:
     """
     Search using Jina's Search API and return the results as a list of SearchResult objects.
     Args:
@@ -20,9 +20,7 @@ def search_jina(query: str, count: int) -> list[SearchResult]:
         list[SearchResult]: A list of search results
     """
     jina_search_endpoint = "https://s.jina.ai/"
-    headers = {
-        "Accept": "application/json",
-    }
+    headers = {"Accept": "application/json", "Authorization": f"Bearer {api_key}"}
     url = str(URL(jina_search_endpoint + query))
     response = requests.get(url, headers=headers)
     response.raise_for_status()

backend/open_webui/apps/retrieval/web/testdata/bing.json

@@ -0,0 +1,58 @@
+{
+	"_type": "SearchResponse",
+	"queryContext": {
+		"originalQuery": "Top 10 international results"
+	},
+	"webPages": {
+		"webSearchUrl": "https://www.bing.com/search?q=Top+10+international+results",
+		"totalEstimatedMatches": 687,
+		"value": [
+			{
+				"id": "https://api.bing.microsoft.com/api/v7/#WebPages.0",
+				"name": "2024 Mexican Grand Prix - F1 results and latest standings ... - PlanetF1",
+				"url": "https://www.planetf1.com/news/f1-results-2024-mexican-grand-prix-race-standings",
+				"datePublished": "2024-10-27T00:00:00.0000000",
+				"datePublishedFreshnessText": "1 day ago",
+				"isFamilyFriendly": true,
+				"displayUrl": "https://www.planetf1.com/news/f1-results-2024-mexican-grand-prix-race-standings",
+				"snippet": "Nico Hulkenberg and Pierre Gasly completed the top 10. A full report of the Mexican Grand Prix is available at the bottom of this article. F1 results – 2024 Mexican Grand Prix",
+				"dateLastCrawled": "2024-10-28T07:15:00.0000000Z",
+				"cachedPageUrl": "https://cc.bingj.com/cache.aspx?q=Top+10+international+results&d=916492551782&mkt=en-US&setlang=en-US&w=zBsfaAPyF2tUrHFHr_vFFdUm8sng4g34",
+				"language": "en",
+				"isNavigational": false,
+				"noCache": false
+			},
+			{
+				"id": "https://api.bing.microsoft.com/api/v7/#WebPages.1",
+				"name": "F1 Results Today: HUGE Verstappen penalties cause major title change",
+				"url": "https://www.gpfans.com/en/f1-news/1033512/f1-results-today-mexican-grand-prix-huge-max-verstappen-penalties-cause-major-title-change/",
+				"datePublished": "2024-10-27T00:00:00.0000000",
+				"datePublishedFreshnessText": "1 day ago",
+				"isFamilyFriendly": true,
+				"displayUrl": "https://www.gpfans.com/en/f1-news/1033512/f1-results-today-mexican-grand-prix-huge-max...",
+				"snippet": "Elsewhere, Mercedes duo Lewis Hamilton and George Russell came home in P4 and P5 respectively. Meanwhile, the surprise package of the day were Haas, with both Kevin Magnussen and Nico Hulkenberg finishing inside the points.. READ MORE: RB star issues apology after red flag CRASH at Mexican GP Mexican Grand Prix 2024 results. 1. Carlos Sainz [Ferrari] 2. Lando Norris [McLaren] - +4.705",
+				"dateLastCrawled": "2024-10-28T06:06:00.0000000Z",
+				"cachedPageUrl": "https://cc.bingj.com/cache.aspx?q=Top+10+international+results&d=2840656522642&mkt=en-US&setlang=en-US&w=-Tbkwxnq52jZCvG7l3CtgcwT1vwAjIUD",
+				"language": "en",
+				"isNavigational": false,
+				"noCache": false
+			},
+			{
+				"id": "https://api.bing.microsoft.com/api/v7/#WebPages.2",
+				"name": "International Power Rankings: England flying, Kangaroos cruising, Fiji rise",
+				"url": "https://www.loverugbyleague.com/post/international-power-rankings-england-flying-kangaroos-cruising-fiji-rise",
+				"datePublished": "2024-10-28T00:00:00.0000000",
+				"datePublishedFreshnessText": "7 hours ago",
+				"isFamilyFriendly": true,
+				"displayUrl": "https://www.loverugbyleague.com/post/international-power-rankings-england-flying...",
+				"snippet": "LRL RECOMMENDS: England player ratings from first Test against Samoa as omnificent George Williams scores perfect 10. 2. Australia (Men) – SAME. The Kangaroos remain 2nd in our Power Rankings after their 22-10 win against New Zealand in Christchurch on Sunday. As was the case in their win against Tonga last week, Mal Meninga’s side weren ...",
+				"dateLastCrawled": "2024-10-28T07:09:00.0000000Z",
+				"cachedPageUrl": "https://cc.bingj.com/cache.aspx?q=Top+10+international+results&d=1535008462672&mkt=en-US&setlang=en-US&w=82ujhH4Kp0iuhCS7wh1xLUFYUeetaVVm",
+				"language": "en",
+				"isNavigational": false,
+				"noCache": false
+			}
+		],
+		"someResultsRemoved": true
+	}
+}

backend/open_webui/apps/socket/main.py

@@ -1,3 +1,5 @@
+# TODO: move socket to webui app
+
 import asyncio
 import socketio
 import logging

backend/open_webui/apps/webui/main.py

@@ -12,6 +12,7 @@ from open_webui.apps.webui.routers import (
     chats,
     folders,
     configs,
+    groups,
     files,
     functions,
     memories,
@@ -34,6 +35,7 @@ from open_webui.config import (
     ENABLE_LOGIN_FORM,
     ENABLE_MESSAGE_RATING,
     ENABLE_SIGNUP,
+    ENABLE_API_KEY,
     ENABLE_EVALUATION_ARENA_MODELS,
     EVALUATION_ARENA_MODELS,
     DEFAULT_ARENA_MODEL,
@@ -50,9 +52,22 @@ from open_webui.config import (
     WEBHOOK_URL,
     WEBUI_AUTH,
     WEBUI_BANNERS,
+    ENABLE_LDAP,
+    LDAP_SERVER_LABEL,
+    LDAP_SERVER_HOST,
+    LDAP_SERVER_PORT,
+    LDAP_ATTRIBUTE_FOR_USERNAME,
+    LDAP_SEARCH_FILTERS,
+    LDAP_SEARCH_BASE,
+    LDAP_APP_DN,
+    LDAP_APP_PASSWORD,
+    LDAP_USE_TLS,
+    LDAP_CA_CERT_FILE,
+    LDAP_CIPHERS,
     AppConfig,
 )
 from open_webui.env import (
+    ENV,
     WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
     WEBUI_AUTH_TRUSTED_NAME_HEADER,
 )
@@ -72,7 +87,11 @@ from open_webui.utils.payload import (
 
 from open_webui.utils.tools import get_tools
 
-app = FastAPI()
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None,
+    openapi_url="/openapi.json" if ENV == "dev" else None,
+    redoc_url=None,
+)
 
 log = logging.getLogger(__name__)
 
@@ -80,6 +99,8 @@ app.state.config = AppConfig()
 
 app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP
 app.state.config.ENABLE_LOGIN_FORM = ENABLE_LOGIN_FORM
+app.state.config.ENABLE_API_KEY = ENABLE_API_KEY
+
 app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN
 app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER
 app.state.AUTH_TRUSTED_NAME_HEADER = WEBUI_AUTH_TRUSTED_NAME_HEADER
@@ -92,6 +113,8 @@ app.state.config.ADMIN_EMAIL = ADMIN_EMAIL
 app.state.config.DEFAULT_MODELS = DEFAULT_MODELS
 app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS
 app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE
+
+
 app.state.config.USER_PERMISSIONS = USER_PERMISSIONS
 app.state.config.WEBHOOK_URL = WEBHOOK_URL
 app.state.config.BANNERS = WEBUI_BANNERS
@@ -111,7 +134,19 @@ app.state.config.OAUTH_ROLES_CLAIM = OAUTH_ROLES_CLAIM
 app.state.config.OAUTH_ALLOWED_ROLES = OAUTH_ALLOWED_ROLES
 app.state.config.OAUTH_ADMIN_ROLES = OAUTH_ADMIN_ROLES
 
-app.state.MODELS = {}
+app.state.config.ENABLE_LDAP = ENABLE_LDAP
+app.state.config.LDAP_SERVER_LABEL = LDAP_SERVER_LABEL
+app.state.config.LDAP_SERVER_HOST = LDAP_SERVER_HOST
+app.state.config.LDAP_SERVER_PORT = LDAP_SERVER_PORT
+app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME = LDAP_ATTRIBUTE_FOR_USERNAME
+app.state.config.LDAP_APP_DN = LDAP_APP_DN
+app.state.config.LDAP_APP_PASSWORD = LDAP_APP_PASSWORD
+app.state.config.LDAP_SEARCH_BASE = LDAP_SEARCH_BASE
+app.state.config.LDAP_SEARCH_FILTERS = LDAP_SEARCH_FILTERS
+app.state.config.LDAP_USE_TLS = LDAP_USE_TLS
+app.state.config.LDAP_CA_CERT_FILE = LDAP_CA_CERT_FILE
+app.state.config.LDAP_CIPHERS = LDAP_CIPHERS
+
 app.state.TOOLS = {}
 app.state.FUNCTIONS = {}
 
@@ -135,13 +170,15 @@ app.include_router(models.router, prefix="/models", tags=["models"])
 app.include_router(knowledge.router, prefix="/knowledge", tags=["knowledge"])
 app.include_router(prompts.router, prefix="/prompts", tags=["prompts"])
 app.include_router(tools.router, prefix="/tools", tags=["tools"])
-app.include_router(functions.router, prefix="/functions", tags=["functions"])
 
 app.include_router(memories.router, prefix="/memories", tags=["memories"])
-app.include_router(evaluations.router, prefix="/evaluations", tags=["evaluations"])
-
 app.include_router(folders.router, prefix="/folders", tags=["folders"])
+
+app.include_router(groups.router, prefix="/groups", tags=["groups"])
 app.include_router(files.router, prefix="/files", tags=["files"])
+app.include_router(functions.router, prefix="/functions", tags=["functions"])
+app.include_router(evaluations.router, prefix="/evaluations", tags=["evaluations"])
+
 
 app.include_router(utils.router, prefix="/utils", tags=["utils"])
 
@@ -336,7 +373,7 @@ def get_function_params(function_module, form_data, user, extra_params=None):
     return params
 
 
-async def generate_function_chat_completion(form_data, user):
+async def generate_function_chat_completion(form_data, user, models: dict = {}):
     model_id = form_data.get("model")
     model_info = Models.get_model_by_id(model_id)
 
@@ -372,6 +409,7 @@ async def generate_function_chat_completion(form_data, user):
             "name": user.name,
             "role": user.role,
         },
+        "__metadata__": metadata,
     }
     extra_params["__tools__"] = get_tools(
         app,
@@ -379,7 +417,7 @@ async def generate_function_chat_completion(form_data, user):
         user,
         {
             **extra_params,
-            "__model__": app.state.MODELS[form_data["model"]],
+            "__model__": models.get(form_data["model"], None),
             "__messages__": form_data["messages"],
             "__files__": files,
         },

backend/open_webui/apps/webui/models/auths.py

@@ -64,6 +64,11 @@ class SigninForm(BaseModel):
     password: str
 
 
+class LdapForm(BaseModel):
+    user: str
+    password: str
+
+
 class ProfileImageUrlForm(BaseModel):
     profile_image_url: str
 

backend/open_webui/apps/webui/models/chats.py

@@ -203,15 +203,22 @@ class ChatTable:
     def update_shared_chat_by_chat_id(self, chat_id: str) -> Optional[ChatModel]:
         try:
             with get_db() as db:
-                print("update_shared_chat_by_id")
                 chat = db.get(Chat, chat_id)
-                print(chat)
-                chat.title = chat.title
-                chat.chat = chat.chat
+                shared_chat = (
+                    db.query(Chat).filter_by(user_id=f"shared-{chat_id}").first()
+                )
+
+                if shared_chat is None:
+                    return self.insert_shared_chat_by_chat_id(chat_id)
+
+                shared_chat.title = chat.title
+                shared_chat.chat = chat.chat
+
+                shared_chat.updated_at = int(time.time())
                 db.commit()
-                db.refresh(chat)
+                db.refresh(shared_chat)
 
-                return self.get_chat_by_id(chat.share_id)
+                return ChatModel.model_validate(shared_chat)
         except Exception:
             return None
 

backend/open_webui/apps/webui/models/documents.py

@@ -1,157 +0,0 @@
-import json
-import logging
-import time
-from typing import Optional
-
-from open_webui.apps.webui.internal.db import Base, get_db
-from open_webui.env import SRC_LOG_LEVELS
-from pydantic import BaseModel, ConfigDict
-from sqlalchemy import BigInteger, Column, String, Text
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Documents DB Schema
-####################
-
-
-class Document(Base):
-    __tablename__ = "document"
-
-    collection_name = Column(String, primary_key=True)
-    name = Column(String, unique=True)
-    title = Column(Text)
-    filename = Column(Text)
-    content = Column(Text, nullable=True)
-    user_id = Column(String)
-    timestamp = Column(BigInteger)
-
-
-class DocumentModel(BaseModel):
-    model_config = ConfigDict(from_attributes=True)
-
-    collection_name: str
-    name: str
-    title: str
-    filename: str
-    content: Optional[str] = None
-    user_id: str
-    timestamp: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class DocumentResponse(BaseModel):
-    collection_name: str
-    name: str
-    title: str
-    filename: str
-    content: Optional[dict] = None
-    user_id: str
-    timestamp: int  # timestamp in epoch
-
-
-class DocumentUpdateForm(BaseModel):
-    name: str
-    title: str
-
-
-class DocumentForm(DocumentUpdateForm):
-    collection_name: str
-    filename: str
-    content: Optional[str] = None
-
-
-class DocumentsTable:
-    def insert_new_doc(
-        self, user_id: str, form_data: DocumentForm
-    ) -> Optional[DocumentModel]:
-        with get_db() as db:
-            document = DocumentModel(
-                **{
-                    **form_data.model_dump(),
-                    "user_id": user_id,
-                    "timestamp": int(time.time()),
-                }
-            )
-
-            try:
-                result = Document(**document.model_dump())
-                db.add(result)
-                db.commit()
-                db.refresh(result)
-                if result:
-                    return DocumentModel.model_validate(result)
-                else:
-                    return None
-            except Exception:
-                return None
-
-    def get_doc_by_name(self, name: str) -> Optional[DocumentModel]:
-        try:
-            with get_db() as db:
-                document = db.query(Document).filter_by(name=name).first()
-                return DocumentModel.model_validate(document) if document else None
-        except Exception:
-            return None
-
-    def get_docs(self) -> list[DocumentModel]:
-        with get_db() as db:
-            return [
-                DocumentModel.model_validate(doc) for doc in db.query(Document).all()
-            ]
-
-    def update_doc_by_name(
-        self, name: str, form_data: DocumentUpdateForm
-    ) -> Optional[DocumentModel]:
-        try:
-            with get_db() as db:
-                db.query(Document).filter_by(name=name).update(
-                    {
-                        "title": form_data.title,
-                        "name": form_data.name,
-                        "timestamp": int(time.time()),
-                    }
-                )
-                db.commit()
-                return self.get_doc_by_name(form_data.name)
-        except Exception as e:
-            log.exception(e)
-            return None
-
-    def update_doc_content_by_name(
-        self, name: str, updated: dict
-    ) -> Optional[DocumentModel]:
-        try:
-            doc = self.get_doc_by_name(name)
-            doc_content = json.loads(doc.content if doc.content else "{}")
-            doc_content = {**doc_content, **updated}
-
-            with get_db() as db:
-                db.query(Document).filter_by(name=name).update(
-                    {
-                        "content": json.dumps(doc_content),
-                        "timestamp": int(time.time()),
-                    }
-                )
-                db.commit()
-                return self.get_doc_by_name(name)
-        except Exception as e:
-            log.exception(e)
-            return None
-
-    def delete_doc_by_name(self, name: str) -> bool:
-        try:
-            with get_db() as db:
-                db.query(Document).filter_by(name=name).delete()
-                db.commit()
-                return True
-        except Exception:
-            return False
-
-
-Documents = DocumentsTable()

backend/open_webui/apps/webui/models/groups.py

@@ -0,0 +1,186 @@
+import json
+import logging
+import time
+from typing import Optional
+import uuid
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.env import SRC_LOG_LEVELS
+
+from open_webui.apps.webui.models.files import FileMetadataResponse
+
+
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text, JSON, func
+
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# UserGroup DB Schema
+####################
+
+
+class Group(Base):
+    __tablename__ = "group"
+
+    id = Column(Text, unique=True, primary_key=True)
+    user_id = Column(Text)
+
+    name = Column(Text)
+    description = Column(Text)
+
+    data = Column(JSON, nullable=True)
+    meta = Column(JSON, nullable=True)
+
+    permissions = Column(JSON, nullable=True)
+    user_ids = Column(JSON, nullable=True)
+
+    created_at = Column(BigInteger)
+    updated_at = Column(BigInteger)
+
+
+class GroupModel(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+    id: str
+    user_id: str
+
+    name: str
+    description: str
+
+    data: Optional[dict] = None
+    meta: Optional[dict] = None
+
+    permissions: Optional[dict] = None
+    user_ids: list[str] = []
+
+    created_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+
+
+####################
+# Forms
+####################
+
+
+class GroupResponse(BaseModel):
+    id: str
+    user_id: str
+    name: str
+    description: str
+    permissions: Optional[dict] = None
+    data: Optional[dict] = None
+    meta: Optional[dict] = None
+    user_ids: list[str] = []
+    created_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+
+
+class GroupForm(BaseModel):
+    name: str
+    description: str
+
+
+class GroupUpdateForm(GroupForm):
+    permissions: Optional[dict] = None
+    user_ids: Optional[list[str]] = None
+    admin_ids: Optional[list[str]] = None
+
+
+class GroupTable:
+    def insert_new_group(
+        self, user_id: str, form_data: GroupForm
+    ) -> Optional[GroupModel]:
+        with get_db() as db:
+            group = GroupModel(
+                **{
+                    **form_data.model_dump(),
+                    "id": str(uuid.uuid4()),
+                    "user_id": user_id,
+                    "created_at": int(time.time()),
+                    "updated_at": int(time.time()),
+                }
+            )
+
+            try:
+                result = Group(**group.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return GroupModel.model_validate(result)
+                else:
+                    return None
+
+            except Exception:
+                return None
+
+    def get_groups(self) -> list[GroupModel]:
+        with get_db() as db:
+            return [
+                GroupModel.model_validate(group)
+                for group in db.query(Group).order_by(Group.updated_at.desc()).all()
+            ]
+
+    def get_groups_by_member_id(self, user_id: str) -> list[GroupModel]:
+        with get_db() as db:
+            return [
+                GroupModel.model_validate(group)
+                for group in db.query(Group)
+                .filter(
+                    func.json_array_length(Group.user_ids) > 0
+                )  # Ensure array exists
+                .filter(
+                    Group.user_ids.cast(String).like(f'%"{user_id}"%')
+                )  # String-based check
+                .order_by(Group.updated_at.desc())
+                .all()
+            ]
+
+    def get_group_by_id(self, id: str) -> Optional[GroupModel]:
+        try:
+            with get_db() as db:
+                group = db.query(Group).filter_by(id=id).first()
+                return GroupModel.model_validate(group) if group else None
+        except Exception:
+            return None
+
+    def update_group_by_id(
+        self, id: str, form_data: GroupUpdateForm, overwrite: bool = False
+    ) -> Optional[GroupModel]:
+        try:
+            with get_db() as db:
+                db.query(Group).filter_by(id=id).update(
+                    {
+                        **form_data.model_dump(exclude_none=True),
+                        "updated_at": int(time.time()),
+                    }
+                )
+                db.commit()
+                return self.get_group_by_id(id=id)
+        except Exception as e:
+            log.exception(e)
+            return None
+
+    def delete_group_by_id(self, id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Group).filter_by(id=id).delete()
+                db.commit()
+                return True
+        except Exception:
+            return False
+
+    def delete_all_groups(self) -> bool:
+        with get_db() as db:
+            try:
+                db.query(Group).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+
+Groups = GroupTable()

backend/open_webui/apps/webui/models/knowledge.py

@@ -8,11 +8,13 @@ from open_webui.apps.webui.internal.db import Base, get_db
 from open_webui.env import SRC_LOG_LEVELS
 
 from open_webui.apps.webui.models.files import FileMetadataResponse
+from open_webui.apps.webui.models.users import Users, UserResponse
 
 
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Column, String, Text, JSON
 
+from open_webui.utils.access_control import has_access
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -34,6 +36,23 @@ class Knowledge(Base):
     data = Column(JSON, nullable=True)
     meta = Column(JSON, nullable=True)
 
+    access_control = Column(JSON, nullable=True)  # Controls data access levels.
+    # Defines access control rules for this entry.
+    # - `None`: Public access, available to all users with the "user" role.
+    # - `{}`: Private access, restricted exclusively to the owner.
+    # - Custom permissions: Specific access control for reading and writing;
+    #   Can specify group or user-level restrictions:
+    #   {
+    #      "read": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      },
+    #      "write": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      }
+    #   }
+
     created_at = Column(BigInteger)
     updated_at = Column(BigInteger)
 
@@ -50,6 +69,8 @@ class KnowledgeModel(BaseModel):
     data: Optional[dict] = None
     meta: Optional[dict] = None
 
+    access_control: Optional[dict] = None
+
     created_at: int  # timestamp in epoch
     updated_at: int  # timestamp in epoch
 
@@ -59,15 +80,15 @@ class KnowledgeModel(BaseModel):
 ####################
 
 
-class KnowledgeResponse(BaseModel):
-    id: str
-    name: str
-    description: str
-    data: Optional[dict] = None
-    meta: Optional[dict] = None
-    created_at: int  # timestamp in epoch
-    updated_at: int  # timestamp in epoch
+class KnowledgeUserModel(KnowledgeModel):
+    user: Optional[UserResponse] = None
+
+
+class KnowledgeResponse(KnowledgeModel):
+    files: Optional[list[FileMetadataResponse | dict]] = None
 
+
+class KnowledgeUserResponse(KnowledgeUserModel):
     files: Optional[list[FileMetadataResponse | dict]] = None
 
 
@@ -75,12 +96,7 @@ class KnowledgeForm(BaseModel):
     name: str
     description: str
     data: Optional[dict] = None
-
-
-class KnowledgeUpdateForm(BaseModel):
-    name: Optional[str] = None
-    description: Optional[str] = None
-    data: Optional[dict] = None
+    access_control: Optional[dict] = None
 
 
 class KnowledgeTable:
@@ -110,14 +126,33 @@ class KnowledgeTable:
             except Exception:
                 return None
 
-    def get_knowledge_items(self) -> list[KnowledgeModel]:
+    def get_knowledge_bases(self) -> list[KnowledgeUserModel]:
         with get_db() as db:
-            return [
-                KnowledgeModel.model_validate(knowledge)
-                for knowledge in db.query(Knowledge)
-                .order_by(Knowledge.updated_at.desc())
-                .all()
-            ]
+            knowledge_bases = []
+            for knowledge in (
+                db.query(Knowledge).order_by(Knowledge.updated_at.desc()).all()
+            ):
+                user = Users.get_user_by_id(knowledge.user_id)
+                knowledge_bases.append(
+                    KnowledgeUserModel.model_validate(
+                        {
+                            **KnowledgeModel.model_validate(knowledge).model_dump(),
+                            "user": user.model_dump() if user else None,
+                        }
+                    )
+                )
+            return knowledge_bases
+
+    def get_knowledge_bases_by_user_id(
+        self, user_id: str, permission: str = "write"
+    ) -> list[KnowledgeUserModel]:
+        knowledge_bases = self.get_knowledge_bases()
+        return [
+            knowledge_base
+            for knowledge_base in knowledge_bases
+            if knowledge_base.user_id == user_id
+            or has_access(user_id, permission, knowledge_base.access_control)
+        ]
 
     def get_knowledge_by_id(self, id: str) -> Optional[KnowledgeModel]:
         try:
@@ -128,14 +163,32 @@ class KnowledgeTable:
             return None
 
     def update_knowledge_by_id(
-        self, id: str, form_data: KnowledgeUpdateForm, overwrite: bool = False
+        self, id: str, form_data: KnowledgeForm, overwrite: bool = False
+    ) -> Optional[KnowledgeModel]:
+        try:
+            with get_db() as db:
+                knowledge = self.get_knowledge_by_id(id=id)
+                db.query(Knowledge).filter_by(id=id).update(
+                    {
+                        **form_data.model_dump(),
+                        "updated_at": int(time.time()),
+                    }
+                )
+                db.commit()
+                return self.get_knowledge_by_id(id=id)
+        except Exception as e:
+            log.exception(e)
+            return None
+
+    def update_knowledge_data_by_id(
+        self, id: str, data: dict
     ) -> Optional[KnowledgeModel]:
         try:
             with get_db() as db:
                 knowledge = self.get_knowledge_by_id(id=id)
                 db.query(Knowledge).filter_by(id=id).update(
                     {
-                        **form_data.model_dump(exclude_none=True),
+                        "data": data,
                         "updated_at": int(time.time()),
                     }
                 )

backend/open_webui/apps/webui/models/models.py

@@ -4,8 +4,19 @@ from typing import Optional
 
 from open_webui.apps.webui.internal.db import Base, JSONField, get_db
 from open_webui.env import SRC_LOG_LEVELS
+
+from open_webui.apps.webui.models.users import Users, UserResponse
+
+
 from pydantic import BaseModel, ConfigDict
-from sqlalchemy import BigInteger, Column, Text
+
+from sqlalchemy import or_, and_, func
+from sqlalchemy.dialects import postgresql, sqlite
+from sqlalchemy import BigInteger, Column, Text, JSON, Boolean
+
+
+from open_webui.utils.access_control import has_access
+
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -67,6 +78,25 @@ class Model(Base):
         Holds a JSON encoded blob of metadata, see `ModelMeta`.
     """
 
+    access_control = Column(JSON, nullable=True)  # Controls data access levels.
+    # Defines access control rules for this entry.
+    # - `None`: Public access, available to all users with the "user" role.
+    # - `{}`: Private access, restricted exclusively to the owner.
+    # - Custom permissions: Specific access control for reading and writing;
+    #   Can specify group or user-level restrictions:
+    #   {
+    #      "read": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      },
+    #      "write": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      }
+    #   }
+
+    is_active = Column(Boolean, default=True)
+
     updated_at = Column(BigInteger)
     created_at = Column(BigInteger)
 
@@ -80,6 +110,9 @@ class ModelModel(BaseModel):
     params: ModelParams
     meta: ModelMeta
 
+    access_control: Optional[dict] = None
+
+    is_active: bool
     updated_at: int  # timestamp in epoch
     created_at: int  # timestamp in epoch
 
@@ -91,12 +124,12 @@ class ModelModel(BaseModel):
 ####################
 
 
-class ModelResponse(BaseModel):
-    id: str
-    name: str
-    meta: ModelMeta
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
+class ModelUserResponse(ModelModel):
+    user: Optional[UserResponse] = None
+
+
+class ModelResponse(ModelModel):
+    pass
 
 
 class ModelForm(BaseModel):
@@ -105,6 +138,8 @@ class ModelForm(BaseModel):
     name: str
     meta: ModelMeta
     params: ModelParams
+    access_control: Optional[dict] = None
+    is_active: bool = True
 
 
 class ModelsTable:
@@ -138,6 +173,39 @@ class ModelsTable:
         with get_db() as db:
             return [ModelModel.model_validate(model) for model in db.query(Model).all()]
 
+    def get_models(self) -> list[ModelUserResponse]:
+        with get_db() as db:
+            models = []
+            for model in db.query(Model).filter(Model.base_model_id != None).all():
+                user = Users.get_user_by_id(model.user_id)
+                models.append(
+                    ModelUserResponse.model_validate(
+                        {
+                            **ModelModel.model_validate(model).model_dump(),
+                            "user": user.model_dump() if user else None,
+                        }
+                    )
+                )
+            return models
+
+    def get_base_models(self) -> list[ModelModel]:
+        with get_db() as db:
+            return [
+                ModelModel.model_validate(model)
+                for model in db.query(Model).filter(Model.base_model_id == None).all()
+            ]
+
+    def get_models_by_user_id(
+        self, user_id: str, permission: str = "write"
+    ) -> list[ModelUserResponse]:
+        models = self.get_models()
+        return [
+            model
+            for model in models
+            if model.user_id == user_id
+            or has_access(user_id, permission, model.access_control)
+        ]
+
     def get_model_by_id(self, id: str) -> Optional[ModelModel]:
         try:
             with get_db() as db:
@@ -146,6 +214,23 @@ class ModelsTable:
         except Exception:
             return None
 
+    def toggle_model_by_id(self, id: str) -> Optional[ModelModel]:
+        with get_db() as db:
+            try:
+                is_active = db.query(Model).filter_by(id=id).first().is_active
+
+                db.query(Model).filter_by(id=id).update(
+                    {
+                        "is_active": not is_active,
+                        "updated_at": int(time.time()),
+                    }
+                )
+                db.commit()
+
+                return self.get_model_by_id(id)
+            except Exception:
+                return None
+
     def update_model_by_id(self, id: str, model: ModelForm) -> Optional[ModelModel]:
         try:
             with get_db() as db:
@@ -153,7 +238,7 @@ class ModelsTable:
                 result = (
                     db.query(Model)
                     .filter_by(id=id)
-                    .update(model.model_dump(exclude={"id"}, exclude_none=True))
+                    .update(model.model_dump(exclude={"id"}))
                 )
                 db.commit()
 
@@ -175,5 +260,15 @@ class ModelsTable:
         except Exception:
             return False
 
+    def delete_all_models(self) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Model).delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
 
 Models = ModelsTable()

backend/open_webui/apps/webui/models/prompts.py

@@ -2,8 +2,12 @@ import time
 from typing import Optional
 
 from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.apps.webui.models.users import Users, UserResponse
+
 from pydantic import BaseModel, ConfigDict
-from sqlalchemy import BigInteger, Column, String, Text
+from sqlalchemy import BigInteger, Column, String, Text, JSON
+
+from open_webui.utils.access_control import has_access
 
 ####################
 # Prompts DB Schema
@@ -19,6 +23,23 @@ class Prompt(Base):
     content = Column(Text)
     timestamp = Column(BigInteger)
 
+    access_control = Column(JSON, nullable=True)  # Controls data access levels.
+    # Defines access control rules for this entry.
+    # - `None`: Public access, available to all users with the "user" role.
+    # - `{}`: Private access, restricted exclusively to the owner.
+    # - Custom permissions: Specific access control for reading and writing;
+    #   Can specify group or user-level restrictions:
+    #   {
+    #      "read": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      },
+    #      "write": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      }
+    #   }
+
 
 class PromptModel(BaseModel):
     command: str
@@ -27,6 +48,7 @@ class PromptModel(BaseModel):
     content: str
     timestamp: int  # timestamp in epoch
 
+    access_control: Optional[dict] = None
     model_config = ConfigDict(from_attributes=True)
 
 
@@ -35,10 +57,15 @@ class PromptModel(BaseModel):
 ####################
 
 
+class PromptUserResponse(PromptModel):
+    user: Optional[UserResponse] = None
+
+
 class PromptForm(BaseModel):
     command: str
     title: str
     content: str
+    access_control: Optional[dict] = None
 
 
 class PromptsTable:
@@ -48,16 +75,14 @@ class PromptsTable:
         prompt = PromptModel(
             **{
                 "user_id": user_id,
-                "command": form_data.command,
-                "title": form_data.title,
-                "content": form_data.content,
+                **form_data.model_dump(),
                 "timestamp": int(time.time()),
             }
         )
 
         try:
             with get_db() as db:
-                result = Prompt(**prompt.dict())
+                result = Prompt(**prompt.model_dump())
                 db.add(result)
                 db.commit()
                 db.refresh(result)
@@ -76,11 +101,34 @@ class PromptsTable:
         except Exception:
             return None
 
-    def get_prompts(self) -> list[PromptModel]:
+    def get_prompts(self) -> list[PromptUserResponse]:
         with get_db() as db:
-            return [
-                PromptModel.model_validate(prompt) for prompt in db.query(Prompt).all()
-            ]
+            prompts = []
+
+            for prompt in db.query(Prompt).order_by(Prompt.timestamp.desc()).all():
+                user = Users.get_user_by_id(prompt.user_id)
+                prompts.append(
+                    PromptUserResponse.model_validate(
+                        {
+                            **PromptModel.model_validate(prompt).model_dump(),
+                            "user": user.model_dump() if user else None,
+                        }
+                    )
+                )
+
+            return prompts
+
+    def get_prompts_by_user_id(
+        self, user_id: str, permission: str = "write"
+    ) -> list[PromptUserResponse]:
+        prompts = self.get_prompts()
+
+        return [
+            prompt
+            for prompt in prompts
+            if prompt.user_id == user_id
+            or has_access(user_id, permission, prompt.access_control)
+        ]
 
     def update_prompt_by_command(
         self, command: str, form_data: PromptForm
@@ -90,6 +138,7 @@ class PromptsTable:
                 prompt = db.query(Prompt).filter_by(command=command).first()
                 prompt.title = form_data.title
                 prompt.content = form_data.content
+                prompt.access_control = form_data.access_control
                 prompt.timestamp = int(time.time())
                 db.commit()
                 return PromptModel.model_validate(prompt)

backend/open_webui/apps/webui/models/tools.py

@@ -3,10 +3,13 @@ import time
 from typing import Optional
 
 from open_webui.apps.webui.internal.db import Base, JSONField, get_db
-from open_webui.apps.webui.models.users import Users
+from open_webui.apps.webui.models.users import Users, UserResponse
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict
-from sqlalchemy import BigInteger, Column, String, Text
+from sqlalchemy import BigInteger, Column, String, Text, JSON
+
+from open_webui.utils.access_control import has_access
+
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -26,6 +29,24 @@ class Tool(Base):
     specs = Column(JSONField)
     meta = Column(JSONField)
     valves = Column(JSONField)
+
+    access_control = Column(JSON, nullable=True)  # Controls data access levels.
+    # Defines access control rules for this entry.
+    # - `None`: Public access, available to all users with the "user" role.
+    # - `{}`: Private access, restricted exclusively to the owner.
+    # - Custom permissions: Specific access control for reading and writing;
+    #   Can specify group or user-level restrictions:
+    #   {
+    #      "read": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      },
+    #      "write": {
+    #          "group_ids": ["group_id1", "group_id2"],
+    #          "user_ids":  ["user_id1", "user_id2"]
+    #      }
+    #   }
+
     updated_at = Column(BigInteger)
     created_at = Column(BigInteger)
 
@@ -42,6 +63,8 @@ class ToolModel(BaseModel):
     content: str
     specs: list[dict]
     meta: ToolMeta
+    access_control: Optional[dict] = None
+
     updated_at: int  # timestamp in epoch
     created_at: int  # timestamp in epoch
 
@@ -58,15 +81,21 @@ class ToolResponse(BaseModel):
     user_id: str
     name: str
     meta: ToolMeta
+    access_control: Optional[dict] = None
     updated_at: int  # timestamp in epoch
     created_at: int  # timestamp in epoch
 
 
+class ToolUserResponse(ToolResponse):
+    user: Optional[UserResponse] = None
+
+
 class ToolForm(BaseModel):
     id: str
     name: str
     content: str
     meta: ToolMeta
+    access_control: Optional[dict] = None
 
 
 class ToolValves(BaseModel):
@@ -109,9 +138,32 @@ class ToolsTable:
         except Exception:
             return None
 
-    def get_tools(self) -> list[ToolModel]:
+    def get_tools(self) -> list[ToolUserResponse]:
         with get_db() as db:
-            return [ToolModel.model_validate(tool) for tool in db.query(Tool).all()]
+            tools = []
+            for tool in db.query(Tool).order_by(Tool.updated_at.desc()).all():
+                user = Users.get_user_by_id(tool.user_id)
+                tools.append(
+                    ToolUserResponse.model_validate(
+                        {
+                            **ToolModel.model_validate(tool).model_dump(),
+                            "user": user.model_dump() if user else None,
+                        }
+                    )
+                )
+            return tools
+
+    def get_tools_by_user_id(
+        self, user_id: str, permission: str = "write"
+    ) -> list[ToolUserResponse]:
+        tools = self.get_tools()
+
+        return [
+            tool
+            for tool in tools
+            if tool.user_id == user_id
+            or has_access(user_id, permission, tool.access_control)
+        ]
 
     def get_tool_valves_by_id(self, id: str) -> Optional[dict]:
         try:

backend/open_webui/apps/webui/models/users.py

@@ -62,6 +62,14 @@ class UserModel(BaseModel):
 ####################
 
 
+class UserResponse(BaseModel):
+    id: str
+    name: str
+    email: str
+    role: str
+    profile_image_url: str
+
+
 class UserRoleUpdateForm(BaseModel):
     id: str
     role: str

backend/open_webui/apps/webui/routers/auths.py

@@ -2,12 +2,14 @@ import re
 import uuid
 import time
 import datetime
+import logging
 
 from open_webui.apps.webui.models.auths import (
     AddUserForm,
     ApiKey,
     Auths,
     Token,
+    LdapForm,
     SigninForm,
     SigninResponse,
     SignupForm,
@@ -16,13 +18,15 @@ from open_webui.apps.webui.models.auths import (
     UserResponse,
 )
 from open_webui.apps.webui.models.users import Users
-from open_webui.config import WEBUI_AUTH
+
 from open_webui.constants import ERROR_MESSAGES, WEBHOOK_MESSAGES
 from open_webui.env import (
+    WEBUI_AUTH,
     WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
     WEBUI_AUTH_TRUSTED_NAME_HEADER,
     WEBUI_SESSION_COOKIE_SAME_SITE,
     WEBUI_SESSION_COOKIE_SECURE,
+    SRC_LOG_LEVELS,
 )
 from fastapi import APIRouter, Depends, HTTPException, Request, status
 from fastapi.responses import Response
@@ -37,10 +41,19 @@ from open_webui.utils.utils import (
     get_password_hash,
 )
 from open_webui.utils.webhook import post_webhook
-from typing import Optional
+from open_webui.utils.access_control import get_permissions
+
+from typing import Optional, List
+
+from ssl import CERT_REQUIRED, PROTOCOL_TLS
+from ldap3 import Server, Connection, ALL, Tls
+from ldap3.utils.conv import escape_filter_chars
 
 router = APIRouter()
 
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MAIN"])
+
 ############################
 # GetSessionUser
 ############################
@@ -48,6 +61,7 @@ router = APIRouter()
 
 class SessionUserResponse(Token, UserResponse):
     expires_at: Optional[int] = None
+    permissions: Optional[dict] = None
 
 
 @router.get("/", response_model=SessionUserResponse)
@@ -80,6 +94,10 @@ async def get_session_user(
         secure=WEBUI_SESSION_COOKIE_SECURE,
     )
 
+    user_permissions = get_permissions(
+        user.id, request.app.state.config.USER_PERMISSIONS
+    )
+
     return {
         "token": token,
         "token_type": "Bearer",
@@ -89,6 +107,7 @@ async def get_session_user(
         "name": user.name,
         "role": user.role,
         "profile_image_url": user.profile_image_url,
+        "permissions": user_permissions,
     }
 
 
@@ -137,6 +156,140 @@ async def update_password(
         raise HTTPException(400, detail=ERROR_MESSAGES.INVALID_CRED)
 
 
+############################
+# LDAP Authentication
+############################
+@router.post("/ldap", response_model=SigninResponse)
+async def ldap_auth(request: Request, response: Response, form_data: LdapForm):
+    ENABLE_LDAP = request.app.state.config.ENABLE_LDAP
+    LDAP_SERVER_LABEL = request.app.state.config.LDAP_SERVER_LABEL
+    LDAP_SERVER_HOST = request.app.state.config.LDAP_SERVER_HOST
+    LDAP_SERVER_PORT = request.app.state.config.LDAP_SERVER_PORT
+    LDAP_ATTRIBUTE_FOR_USERNAME = request.app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME
+    LDAP_SEARCH_BASE = request.app.state.config.LDAP_SEARCH_BASE
+    LDAP_SEARCH_FILTERS = request.app.state.config.LDAP_SEARCH_FILTERS
+    LDAP_APP_DN = request.app.state.config.LDAP_APP_DN
+    LDAP_APP_PASSWORD = request.app.state.config.LDAP_APP_PASSWORD
+    LDAP_USE_TLS = request.app.state.config.LDAP_USE_TLS
+    LDAP_CA_CERT_FILE = request.app.state.config.LDAP_CA_CERT_FILE
+    LDAP_CIPHERS = (
+        request.app.state.config.LDAP_CIPHERS
+        if request.app.state.config.LDAP_CIPHERS
+        else "ALL"
+    )
+
+    if not ENABLE_LDAP:
+        raise HTTPException(400, detail="LDAP authentication is not enabled")
+
+    try:
+        tls = Tls(
+            validate=CERT_REQUIRED,
+            version=PROTOCOL_TLS,
+            ca_certs_file=LDAP_CA_CERT_FILE,
+            ciphers=LDAP_CIPHERS,
+        )
+    except Exception as e:
+        log.error(f"An error occurred on TLS: {str(e)}")
+        raise HTTPException(400, detail=str(e))
+
+    try:
+        server = Server(
+            host=LDAP_SERVER_HOST,
+            port=LDAP_SERVER_PORT,
+            get_info=ALL,
+            use_ssl=LDAP_USE_TLS,
+            tls=tls,
+        )
+        connection_app = Connection(
+            server,
+            LDAP_APP_DN,
+            LDAP_APP_PASSWORD,
+            auto_bind="NONE",
+            authentication="SIMPLE",
+        )
+        if not connection_app.bind():
+            raise HTTPException(400, detail="Application account bind failed")
+
+        search_success = connection_app.search(
+            search_base=LDAP_SEARCH_BASE,
+            search_filter=f"(&({LDAP_ATTRIBUTE_FOR_USERNAME}={escape_filter_chars(form_data.user.lower())}){LDAP_SEARCH_FILTERS})",
+            attributes=[f"{LDAP_ATTRIBUTE_FOR_USERNAME}", "mail", "cn"],
+        )
+
+        if not search_success:
+            raise HTTPException(400, detail="User not found in the LDAP server")
+
+        entry = connection_app.entries[0]
+        username = str(entry[f"{LDAP_ATTRIBUTE_FOR_USERNAME}"]).lower()
+        mail = str(entry["mail"])
+        cn = str(entry["cn"])
+        user_dn = entry.entry_dn
+
+        if username == form_data.user.lower():
+            connection_user = Connection(
+                server,
+                user_dn,
+                form_data.password,
+                auto_bind="NONE",
+                authentication="SIMPLE",
+            )
+            if not connection_user.bind():
+                raise HTTPException(400, f"Authentication failed for {form_data.user}")
+
+            user = Users.get_user_by_email(mail)
+            if not user:
+
+                try:
+                    hashed = get_password_hash(form_data.password)
+                    user = Auths.insert_new_auth(mail, hashed, cn)
+
+                    if not user:
+                        raise HTTPException(
+                            500, detail=ERROR_MESSAGES.CREATE_USER_ERROR
+                        )
+
+                except HTTPException:
+                    raise
+                except Exception as err:
+                    raise HTTPException(500, detail=ERROR_MESSAGES.DEFAULT(err))
+
+            user = Auths.authenticate_user(mail, password=str(form_data.password))
+
+            if user:
+                token = create_token(
+                    data={"id": user.id},
+                    expires_delta=parse_duration(
+                        request.app.state.config.JWT_EXPIRES_IN
+                    ),
+                )
+
+                # Set the cookie token
+                response.set_cookie(
+                    key="token",
+                    value=token,
+                    httponly=True,  # Ensures the cookie is not accessible via JavaScript
+                )
+
+                return {
+                    "token": token,
+                    "token_type": "Bearer",
+                    "id": user.id,
+                    "email": user.email,
+                    "name": user.name,
+                    "role": user.role,
+                    "profile_image_url": user.profile_image_url,
+                }
+            else:
+                raise HTTPException(400, detail=ERROR_MESSAGES.INVALID_CRED)
+        else:
+            raise HTTPException(
+                400,
+                f"User {form_data.user} does not match the record. Search result: {str(entry[f'{LDAP_ATTRIBUTE_FOR_USERNAME}'])}",
+            )
+    except Exception as e:
+        raise HTTPException(400, detail=str(e))
+
+
 ############################
 # SignIn
 ############################
@@ -211,6 +364,10 @@ async def signin(request: Request, response: Response, form_data: SigninForm):
             secure=WEBUI_SESSION_COOKIE_SECURE,
         )
 
+        user_permissions = get_permissions(
+            user.id, request.app.state.config.USER_PERMISSIONS
+        )
+
         return {
             "token": token,
             "token_type": "Bearer",
@@ -220,6 +377,7 @@ async def signin(request: Request, response: Response, form_data: SigninForm):
             "name": user.name,
             "role": user.role,
             "profile_image_url": user.profile_image_url,
+            "permissions": user_permissions,
         }
     else:
         raise HTTPException(400, detail=ERROR_MESSAGES.INVALID_CRED)
@@ -260,6 +418,11 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
             if Users.get_num_users() == 0
             else request.app.state.config.DEFAULT_USER_ROLE
         )
+
+        if Users.get_num_users() == 0:
+            # Disable signup after the first user is created
+            request.app.state.config.ENABLE_SIGNUP = False
+
         hashed = get_password_hash(form_data.password)
         user = Auths.insert_new_auth(
             form_data.email.lower(),
@@ -307,6 +470,10 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
                     },
                 )
 
+            user_permissions = get_permissions(
+                user.id, request.app.state.config.USER_PERMISSIONS
+            )
+
             return {
                 "token": token,
                 "token_type": "Bearer",
@@ -316,6 +483,7 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
                 "name": user.name,
                 "role": user.role,
                 "profile_image_url": user.profile_image_url,
+                "permissions": user_permissions,
             }
         else:
             raise HTTPException(500, detail=ERROR_MESSAGES.CREATE_USER_ERROR)
@@ -413,6 +581,7 @@ async def get_admin_config(request: Request, user=Depends(get_admin_user)):
     return {
         "SHOW_ADMIN_DETAILS": request.app.state.config.SHOW_ADMIN_DETAILS,
         "ENABLE_SIGNUP": request.app.state.config.ENABLE_SIGNUP,
+        "ENABLE_API_KEY": request.app.state.config.ENABLE_API_KEY,
         "DEFAULT_USER_ROLE": request.app.state.config.DEFAULT_USER_ROLE,
         "JWT_EXPIRES_IN": request.app.state.config.JWT_EXPIRES_IN,
         "ENABLE_COMMUNITY_SHARING": request.app.state.config.ENABLE_COMMUNITY_SHARING,
@@ -423,6 +592,7 @@ async def get_admin_config(request: Request, user=Depends(get_admin_user)):
 class AdminConfig(BaseModel):
     SHOW_ADMIN_DETAILS: bool
     ENABLE_SIGNUP: bool
+    ENABLE_API_KEY: bool
     DEFAULT_USER_ROLE: str
     JWT_EXPIRES_IN: str
     ENABLE_COMMUNITY_SHARING: bool
@@ -435,6 +605,7 @@ async def update_admin_config(
 ):
     request.app.state.config.SHOW_ADMIN_DETAILS = form_data.SHOW_ADMIN_DETAILS
     request.app.state.config.ENABLE_SIGNUP = form_data.ENABLE_SIGNUP
+    request.app.state.config.ENABLE_API_KEY = form_data.ENABLE_API_KEY
 
     if form_data.DEFAULT_USER_ROLE in ["pending", "user", "admin"]:
         request.app.state.config.DEFAULT_USER_ROLE = form_data.DEFAULT_USER_ROLE
@@ -453,6 +624,7 @@ async def update_admin_config(
     return {
         "SHOW_ADMIN_DETAILS": request.app.state.config.SHOW_ADMIN_DETAILS,
         "ENABLE_SIGNUP": request.app.state.config.ENABLE_SIGNUP,
+        "ENABLE_API_KEY": request.app.state.config.ENABLE_API_KEY,
         "DEFAULT_USER_ROLE": request.app.state.config.DEFAULT_USER_ROLE,
         "JWT_EXPIRES_IN": request.app.state.config.JWT_EXPIRES_IN,
         "ENABLE_COMMUNITY_SHARING": request.app.state.config.ENABLE_COMMUNITY_SHARING,
@@ -460,6 +632,105 @@ async def update_admin_config(
     }
 
 
+class LdapServerConfig(BaseModel):
+    label: str
+    host: str
+    port: Optional[int] = None
+    attribute_for_username: str = "uid"
+    app_dn: str
+    app_dn_password: str
+    search_base: str
+    search_filters: str = ""
+    use_tls: bool = True
+    certificate_path: Optional[str] = None
+    ciphers: Optional[str] = "ALL"
+
+
+@router.get("/admin/config/ldap/server", response_model=LdapServerConfig)
+async def get_ldap_server(request: Request, user=Depends(get_admin_user)):
+    return {
+        "label": request.app.state.config.LDAP_SERVER_LABEL,
+        "host": request.app.state.config.LDAP_SERVER_HOST,
+        "port": request.app.state.config.LDAP_SERVER_PORT,
+        "attribute_for_username": request.app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME,
+        "app_dn": request.app.state.config.LDAP_APP_DN,
+        "app_dn_password": request.app.state.config.LDAP_APP_PASSWORD,
+        "search_base": request.app.state.config.LDAP_SEARCH_BASE,
+        "search_filters": request.app.state.config.LDAP_SEARCH_FILTERS,
+        "use_tls": request.app.state.config.LDAP_USE_TLS,
+        "certificate_path": request.app.state.config.LDAP_CA_CERT_FILE,
+        "ciphers": request.app.state.config.LDAP_CIPHERS,
+    }
+
+
+@router.post("/admin/config/ldap/server")
+async def update_ldap_server(
+    request: Request, form_data: LdapServerConfig, user=Depends(get_admin_user)
+):
+    required_fields = [
+        "label",
+        "host",
+        "attribute_for_username",
+        "app_dn",
+        "app_dn_password",
+        "search_base",
+    ]
+    for key in required_fields:
+        value = getattr(form_data, key)
+        if not value:
+            raise HTTPException(400, detail=f"Required field {key} is empty")
+
+    if form_data.use_tls and not form_data.certificate_path:
+        raise HTTPException(
+            400, detail="TLS is enabled but certificate file path is missing"
+        )
+
+    request.app.state.config.LDAP_SERVER_LABEL = form_data.label
+    request.app.state.config.LDAP_SERVER_HOST = form_data.host
+    request.app.state.config.LDAP_SERVER_PORT = form_data.port
+    request.app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME = (
+        form_data.attribute_for_username
+    )
+    request.app.state.config.LDAP_APP_DN = form_data.app_dn
+    request.app.state.config.LDAP_APP_PASSWORD = form_data.app_dn_password
+    request.app.state.config.LDAP_SEARCH_BASE = form_data.search_base
+    request.app.state.config.LDAP_SEARCH_FILTERS = form_data.search_filters
+    request.app.state.config.LDAP_USE_TLS = form_data.use_tls
+    request.app.state.config.LDAP_CA_CERT_FILE = form_data.certificate_path
+    request.app.state.config.LDAP_CIPHERS = form_data.ciphers
+
+    return {
+        "label": request.app.state.config.LDAP_SERVER_LABEL,
+        "host": request.app.state.config.LDAP_SERVER_HOST,
+        "port": request.app.state.config.LDAP_SERVER_PORT,
+        "attribute_for_username": request.app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME,
+        "app_dn": request.app.state.config.LDAP_APP_DN,
+        "app_dn_password": request.app.state.config.LDAP_APP_PASSWORD,
+        "search_base": request.app.state.config.LDAP_SEARCH_BASE,
+        "search_filters": request.app.state.config.LDAP_SEARCH_FILTERS,
+        "use_tls": request.app.state.config.LDAP_USE_TLS,
+        "certificate_path": request.app.state.config.LDAP_CA_CERT_FILE,
+        "ciphers": request.app.state.config.LDAP_CIPHERS,
+    }
+
+
+@router.get("/admin/config/ldap")
+async def get_ldap_config(request: Request, user=Depends(get_admin_user)):
+    return {"ENABLE_LDAP": request.app.state.config.ENABLE_LDAP}
+
+
+class LdapConfigForm(BaseModel):
+    enable_ldap: Optional[bool] = None
+
+
+@router.post("/admin/config/ldap")
+async def update_ldap_config(
+    request: Request, form_data: LdapConfigForm, user=Depends(get_admin_user)
+):
+    request.app.state.config.ENABLE_LDAP = form_data.enable_ldap
+    return {"ENABLE_LDAP": request.app.state.config.ENABLE_LDAP}
+
+
 ############################
 # API Key
 ############################
@@ -467,9 +738,16 @@ async def update_admin_config(
 
 # create api key
 @router.post("/api_key", response_model=ApiKey)
-async def create_api_key_(user=Depends(get_current_user)):
+async def generate_api_key(request: Request, user=Depends(get_current_user)):
+    if not request.app.state.config.ENABLE_API_KEY:
+        raise HTTPException(
+            status.HTTP_403_FORBIDDEN,
+            detail=ERROR_MESSAGES.API_KEY_CREATION_NOT_ALLOWED,
+        )
+
     api_key = create_api_key()
     success = Users.update_user_api_key_by_id(user.id, api_key)
+
     if success:
         return {
             "api_key": api_key,

backend/open_webui/apps/webui/routers/chats.py

@@ -17,7 +17,10 @@ from open_webui.constants import ERROR_MESSAGES
 from open_webui.env import SRC_LOG_LEVELS
 from fastapi import APIRouter, Depends, HTTPException, Request, status
 from pydantic import BaseModel
+
+
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_permission
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -50,9 +53,10 @@ async def get_session_user_chat_list(
 
 @router.delete("/", response_model=bool)
 async def delete_all_user_chats(request: Request, user=Depends(get_verified_user)):
-    if user.role == "user" and not request.app.state.config.USER_PERMISSIONS.get(
-        "chat", {}
-    ).get("deletion", {}):
+
+    if user.role == "user" and not has_permission(
+        user.id, "chat.delete", request.app.state.config.USER_PERMISSIONS
+    ):
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
             detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
@@ -385,8 +389,8 @@ async def delete_chat_by_id(request: Request, id: str, user=Depends(get_verified
 
         return result
     else:
-        if not request.app.state.config.USER_PERMISSIONS.get("chat", {}).get(
-            "deletion", {}
+        if not has_permission(
+            user.id, "chat.delete", request.app.state.config.USER_PERMISSIONS
         ):
             raise HTTPException(
                 status_code=status.HTTP_401_UNAUTHORIZED,

backend/open_webui/apps/webui/routers/groups.py

@@ -0,0 +1,120 @@
+import os
+from pathlib import Path
+from typing import Optional
+
+from open_webui.apps.webui.models.groups import (
+    Groups,
+    GroupForm,
+    GroupUpdateForm,
+    GroupResponse,
+)
+
+from open_webui.config import CACHE_DIR
+from open_webui.constants import ERROR_MESSAGES
+from fastapi import APIRouter, Depends, HTTPException, Request, status
+from open_webui.utils.utils import get_admin_user, get_verified_user
+
+router = APIRouter()
+
+############################
+# GetFunctions
+############################
+
+
+@router.get("/", response_model=list[GroupResponse])
+async def get_groups(user=Depends(get_verified_user)):
+    if user.role == "admin":
+        return Groups.get_groups()
+    else:
+        return Groups.get_groups_by_member_id(user.id)
+
+
+############################
+# CreateNewGroup
+############################
+
+
+@router.post("/create", response_model=Optional[GroupResponse])
+async def create_new_function(form_data: GroupForm, user=Depends(get_admin_user)):
+    try:
+        group = Groups.insert_new_group(user.id, form_data)
+        if group:
+            return group
+        else:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=ERROR_MESSAGES.DEFAULT("Error creating group"),
+            )
+    except Exception as e:
+        print(e)
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+############################
+# GetGroupById
+############################
+
+
+@router.get("/id/{id}", response_model=Optional[GroupResponse])
+async def get_group_by_id(id: str, user=Depends(get_admin_user)):
+    group = Groups.get_group_by_id(id)
+    if group:
+        return group
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+
+############################
+# UpdateGroupById
+############################
+
+
+@router.post("/id/{id}/update", response_model=Optional[GroupResponse])
+async def update_group_by_id(
+    id: str, form_data: GroupUpdateForm, user=Depends(get_admin_user)
+):
+    try:
+        group = Groups.update_group_by_id(id, form_data)
+        if group:
+            return group
+        else:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=ERROR_MESSAGES.DEFAULT("Error updating group"),
+            )
+    except Exception as e:
+        print(e)
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+############################
+# DeleteGroupById
+############################
+
+
+@router.delete("/id/{id}/delete", response_model=bool)
+async def delete_group_by_id(id: str, user=Depends(get_admin_user)):
+    try:
+        result = Groups.delete_group_by_id(id)
+        if result:
+            return result
+        else:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=ERROR_MESSAGES.DEFAULT("Error deleting group"),
+            )
+    except Exception as e:
+        print(e)
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )

backend/open_webui/apps/webui/routers/knowledge.py

@@ -1,14 +1,14 @@
 import json
 from typing import Optional, Union
 from pydantic import BaseModel
-from fastapi import APIRouter, Depends, HTTPException, status
+from fastapi import APIRouter, Depends, HTTPException, status, Request
 import logging
 
 from open_webui.apps.webui.models.knowledge import (
     Knowledges,
-    KnowledgeUpdateForm,
     KnowledgeForm,
     KnowledgeResponse,
+    KnowledgeUserResponse,
 )
 from open_webui.apps.webui.models.files import Files, FileModel
 from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
@@ -17,6 +17,9 @@ from open_webui.apps.retrieval.main import process_file, ProcessFileForm
 
 from open_webui.constants import ERROR_MESSAGES
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access, has_permission
+
+
 from open_webui.env import SRC_LOG_LEVELS
 
 
@@ -26,64 +29,103 @@ log.setLevel(SRC_LOG_LEVELS["MODELS"])
 router = APIRouter()
 
 ############################
-# GetKnowledgeItems
+# getKnowledgeBases
 ############################
 
 
-@router.get(
-    "/", response_model=Optional[Union[list[KnowledgeResponse], KnowledgeResponse]]
-)
-async def get_knowledge_items(
-    id: Optional[str] = None, user=Depends(get_verified_user)
-):
-    if id:
-        knowledge = Knowledges.get_knowledge_by_id(id=id)
+@router.get("/", response_model=list[KnowledgeUserResponse])
+async def get_knowledge(user=Depends(get_verified_user)):
+    knowledge_bases = []
 
-        if knowledge:
-            return knowledge
-        else:
-            raise HTTPException(
-                status_code=status.HTTP_401_UNAUTHORIZED,
-                detail=ERROR_MESSAGES.NOT_FOUND,
-            )
+    if user.role == "admin":
+        knowledge_bases = Knowledges.get_knowledge_bases()
     else:
-        knowledge_bases = []
-
-        for knowledge in Knowledges.get_knowledge_items():
+        knowledge_bases = Knowledges.get_knowledge_bases_by_user_id(user.id, "read")
+
+    # Get files for each knowledge base
+    knowledge_with_files = []
+    for knowledge_base in knowledge_bases:
+        files = []
+        if knowledge_base.data:
+            files = Files.get_file_metadatas_by_ids(
+                knowledge_base.data.get("file_ids", [])
+            )
 
-            files = []
-            if knowledge.data:
-                files = Files.get_file_metadatas_by_ids(
-                    knowledge.data.get("file_ids", [])
+            # Check if all files exist
+            if len(files) != len(knowledge_base.data.get("file_ids", [])):
+                missing_files = list(
+                    set(knowledge_base.data.get("file_ids", []))
+                    - set([file.id for file in files])
                 )
+                if missing_files:
+                    data = knowledge_base.data or {}
+                    file_ids = data.get("file_ids", [])
+
+                    for missing_file in missing_files:
+                        file_ids.remove(missing_file)
 
-                # Check if all files exist
-                if len(files) != len(knowledge.data.get("file_ids", [])):
-                    missing_files = list(
-                        set(knowledge.data.get("file_ids", []))
-                        - set([file.id for file in files])
+                    data["file_ids"] = file_ids
+                    Knowledges.update_knowledge_data_by_id(
+                        id=knowledge_base.id, data=data
                     )
-                    if missing_files:
-                        data = knowledge.data or {}
-                        file_ids = data.get("file_ids", [])
 
-                        for missing_file in missing_files:
-                            file_ids.remove(missing_file)
+                    files = Files.get_file_metadatas_by_ids(file_ids)
 
-                        data["file_ids"] = file_ids
-                        Knowledges.update_knowledge_by_id(
-                            id=knowledge.id, form_data=KnowledgeUpdateForm(data=data)
-                        )
+        knowledge_with_files.append(
+            KnowledgeUserResponse(
+                **knowledge_base.model_dump(),
+                files=files,
+            )
+        )
 
-                        files = Files.get_file_metadatas_by_ids(file_ids)
+    return knowledge_with_files
 
-            knowledge_bases.append(
-                KnowledgeResponse(
-                    **knowledge.model_dump(),
-                    files=files,
+
+@router.get("/list", response_model=list[KnowledgeUserResponse])
+async def get_knowledge_list(user=Depends(get_verified_user)):
+    knowledge_bases = []
+
+    if user.role == "admin":
+        knowledge_bases = Knowledges.get_knowledge_bases()
+    else:
+        knowledge_bases = Knowledges.get_knowledge_bases_by_user_id(user.id, "write")
+
+    # Get files for each knowledge base
+    knowledge_with_files = []
+    for knowledge_base in knowledge_bases:
+        files = []
+        if knowledge_base.data:
+            files = Files.get_file_metadatas_by_ids(
+                knowledge_base.data.get("file_ids", [])
+            )
+
+            # Check if all files exist
+            if len(files) != len(knowledge_base.data.get("file_ids", [])):
+                missing_files = list(
+                    set(knowledge_base.data.get("file_ids", []))
+                    - set([file.id for file in files])
                 )
+                if missing_files:
+                    data = knowledge_base.data or {}
+                    file_ids = data.get("file_ids", [])
+
+                    for missing_file in missing_files:
+                        file_ids.remove(missing_file)
+
+                    data["file_ids"] = file_ids
+                    Knowledges.update_knowledge_data_by_id(
+                        id=knowledge_base.id, data=data
+                    )
+
+                    files = Files.get_file_metadatas_by_ids(file_ids)
+
+        knowledge_with_files.append(
+            KnowledgeUserResponse(
+                **knowledge_base.model_dump(),
+                files=files,
             )
-        return knowledge_bases
+        )
+    return knowledge_with_files
 
 
 ############################
@@ -92,7 +134,17 @@ async def get_knowledge_items(
 
 
 @router.post("/create", response_model=Optional[KnowledgeResponse])
-async def create_new_knowledge(form_data: KnowledgeForm, user=Depends(get_admin_user)):
+async def create_new_knowledge(
+    request: Request, form_data: KnowledgeForm, user=Depends(get_verified_user)
+):
+    if user.role != "admin" and not has_permission(
+        user.id, "workspace.knowledge", request.app.state.config.USER_PERMISSIONS
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
     knowledge = Knowledges.insert_new_knowledge(user.id, form_data)
 
     if knowledge:
@@ -118,13 +170,20 @@ async def get_knowledge_by_id(id: str, user=Depends(get_verified_user)):
     knowledge = Knowledges.get_knowledge_by_id(id=id)
 
     if knowledge:
-        file_ids = knowledge.data.get("file_ids", []) if knowledge.data else []
-        files = Files.get_files_by_ids(file_ids)
 
-        return KnowledgeFilesResponse(
-            **knowledge.model_dump(),
-            files=files,
-        )
+        if (
+            user.role == "admin"
+            or knowledge.user_id == user.id
+            or has_access(user.id, "read", knowledge.access_control)
+        ):
+
+            file_ids = knowledge.data.get("file_ids", []) if knowledge.data else []
+            files = Files.get_files_by_ids(file_ids)
+
+            return KnowledgeFilesResponse(
+                **knowledge.model_dump(),
+                files=files,
+            )
     else:
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
@@ -140,11 +199,23 @@ async def get_knowledge_by_id(id: str, user=Depends(get_verified_user)):
 @router.post("/{id}/update", response_model=Optional[KnowledgeFilesResponse])
 async def update_knowledge_by_id(
     id: str,
-    form_data: KnowledgeUpdateForm,
-    user=Depends(get_admin_user),
+    form_data: KnowledgeForm,
+    user=Depends(get_verified_user),
 ):
-    knowledge = Knowledges.update_knowledge_by_id(id=id, form_data=form_data)
+    knowledge = Knowledges.get_knowledge_by_id(id=id)
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
 
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
+    knowledge = Knowledges.update_knowledge_by_id(id=id, form_data=form_data)
     if knowledge:
         file_ids = knowledge.data.get("file_ids", []) if knowledge.data else []
         files = Files.get_files_by_ids(file_ids)
@@ -173,9 +244,22 @@ class KnowledgeFileIdForm(BaseModel):
 def add_file_to_knowledge_by_id(
     id: str,
     form_data: KnowledgeFileIdForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
     knowledge = Knowledges.get_knowledge_by_id(id=id)
+
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     file = Files.get_file_by_id(form_data.file_id)
     if not file:
         raise HTTPException(
@@ -206,9 +290,7 @@ def add_file_to_knowledge_by_id(
             file_ids.append(form_data.file_id)
             data["file_ids"] = file_ids
 
-            knowledge = Knowledges.update_knowledge_by_id(
-                id=id, form_data=KnowledgeUpdateForm(data=data)
-            )
+            knowledge = Knowledges.update_knowledge_data_by_id(id=id, data=data)
 
             if knowledge:
                 files = Files.get_files_by_ids(file_ids)
@@ -238,9 +320,21 @@ def add_file_to_knowledge_by_id(
 def update_file_from_knowledge_by_id(
     id: str,
     form_data: KnowledgeFileIdForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
     knowledge = Knowledges.get_knowledge_by_id(id=id)
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     file = Files.get_file_by_id(form_data.file_id)
     if not file:
         raise HTTPException(
@@ -288,9 +382,21 @@ def update_file_from_knowledge_by_id(
 def remove_file_from_knowledge_by_id(
     id: str,
     form_data: KnowledgeFileIdForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
     knowledge = Knowledges.get_knowledge_by_id(id=id)
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     file = Files.get_file_by_id(form_data.file_id)
     if not file:
         raise HTTPException(
@@ -318,9 +424,7 @@ def remove_file_from_knowledge_by_id(
             file_ids.remove(form_data.file_id)
             data["file_ids"] = file_ids
 
-            knowledge = Knowledges.update_knowledge_by_id(
-                id=id, form_data=KnowledgeUpdateForm(data=data)
-            )
+            knowledge = Knowledges.update_knowledge_data_by_id(id=id, data=data)
 
             if knowledge:
                 files = Files.get_files_by_ids(file_ids)
@@ -347,35 +451,60 @@ def remove_file_from_knowledge_by_id(
 
 
 ############################
-# ResetKnowledgeById
+# DeleteKnowledgeById
 ############################
 
 
-@router.post("/{id}/reset", response_model=Optional[KnowledgeResponse])
-async def reset_knowledge_by_id(id: str, user=Depends(get_admin_user)):
+@router.delete("/{id}/delete", response_model=bool)
+async def delete_knowledge_by_id(id: str, user=Depends(get_verified_user)):
+    knowledge = Knowledges.get_knowledge_by_id(id=id)
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     try:
         VECTOR_DB_CLIENT.delete_collection(collection_name=id)
     except Exception as e:
         log.debug(e)
         pass
-
-    knowledge = Knowledges.update_knowledge_by_id(
-        id=id, form_data=KnowledgeUpdateForm(data={"file_ids": []})
-    )
-    return knowledge
+    result = Knowledges.delete_knowledge_by_id(id=id)
+    return result
 
 
 ############################
-# DeleteKnowledgeById
+# ResetKnowledgeById
 ############################
 
 
-@router.delete("/{id}/delete", response_model=bool)
-async def delete_knowledge_by_id(id: str, user=Depends(get_admin_user)):
+@router.post("/{id}/reset", response_model=Optional[KnowledgeResponse])
+async def reset_knowledge_by_id(id: str, user=Depends(get_verified_user)):
+    knowledge = Knowledges.get_knowledge_by_id(id=id)
+    if not knowledge:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if knowledge.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     try:
         VECTOR_DB_CLIENT.delete_collection(collection_name=id)
     except Exception as e:
         log.debug(e)
         pass
-    result = Knowledges.delete_knowledge_by_id(id=id)
-    return result
+
+    knowledge = Knowledges.update_knowledge_data_by_id(id=id, data={"file_ids": []})
+
+    return knowledge

backend/open_webui/apps/webui/routers/models.py

@@ -4,53 +4,71 @@ from open_webui.apps.webui.models.models import (
     ModelForm,
     ModelModel,
     ModelResponse,
+    ModelUserResponse,
     Models,
 )
 from open_webui.constants import ERROR_MESSAGES
 from fastapi import APIRouter, Depends, HTTPException, Request, status
+
+
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access, has_permission
+
 
 router = APIRouter()
 
+
 ###########################
-# getModels
+# GetModels
 ###########################
 
 
-@router.get("/", response_model=list[ModelResponse])
+@router.get("/", response_model=list[ModelUserResponse])
 async def get_models(id: Optional[str] = None, user=Depends(get_verified_user)):
-    if id:
-        model = Models.get_model_by_id(id)
-        if model:
-            return [model]
-        else:
-            raise HTTPException(
-                status_code=status.HTTP_401_UNAUTHORIZED,
-                detail=ERROR_MESSAGES.NOT_FOUND,
-            )
+    if user.role == "admin":
+        return Models.get_models()
     else:
-        return Models.get_all_models()
+        return Models.get_models_by_user_id(user.id)
+
+
+###########################
+# GetBaseModels
+###########################
+
+
+@router.get("/base", response_model=list[ModelResponse])
+async def get_base_models(user=Depends(get_admin_user)):
+    return Models.get_base_models()
 
 
 ############################
-# AddNewModel
+# CreateNewModel
 ############################
 
 
-@router.post("/add", response_model=Optional[ModelModel])
-async def add_new_model(
+@router.post("/create", response_model=Optional[ModelModel])
+async def create_new_model(
     request: Request,
     form_data: ModelForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
-    if form_data.id in request.app.state.MODELS:
+    if user.role != "admin" and not has_permission(
+        user.id, "workspace.models", request.app.state.config.USER_PERMISSIONS
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
+    model = Models.get_model_by_id(form_data.id)
+    if model:
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
             detail=ERROR_MESSAGES.MODEL_ID_TAKEN,
         )
+
     else:
         model = Models.insert_new_model(form_data, user.id)
-
         if model:
             return model
         else:
@@ -60,37 +78,85 @@ async def add_new_model(
             )
 
 
+###########################
+# GetModelById
+###########################
+
+
+# Note: We're not using the typical url path param here, but instead using a query parameter to allow '/' in the id
+@router.get("/model", response_model=Optional[ModelResponse])
+async def get_model_by_id(id: str, user=Depends(get_verified_user)):
+    model = Models.get_model_by_id(id)
+    if model:
+        if (
+            user.role == "admin"
+            or model.user_id == user.id
+            or has_access(user.id, "read", model.access_control)
+        ):
+            return model
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+
 ############################
-# UpdateModelById
+# ToggelModelById
 ############################
 
 
-@router.post("/update", response_model=Optional[ModelModel])
-async def update_model_by_id(
-    request: Request,
-    id: str,
-    form_data: ModelForm,
-    user=Depends(get_admin_user),
-):
+@router.post("/model/toggle", response_model=Optional[ModelResponse])
+async def toggle_model_by_id(id: str, user=Depends(get_verified_user)):
     model = Models.get_model_by_id(id)
     if model:
-        model = Models.update_model_by_id(id, form_data)
-        return model
-    else:
-        if form_data.id in request.app.state.MODELS:
-            model = Models.insert_new_model(form_data, user.id)
+        if (
+            user.role == "admin"
+            or model.user_id == user.id
+            or has_access(user.id, "write", model.access_control)
+        ):
+            model = Models.toggle_model_by_id(id)
+
             if model:
                 return model
             else:
                 raise HTTPException(
-                    status_code=status.HTTP_401_UNAUTHORIZED,
-                    detail=ERROR_MESSAGES.DEFAULT(),
+                    status_code=status.HTTP_400_BAD_REQUEST,
+                    detail=ERROR_MESSAGES.DEFAULT("Error updating function"),
                 )
         else:
             raise HTTPException(
                 status_code=status.HTTP_401_UNAUTHORIZED,
-                detail=ERROR_MESSAGES.DEFAULT(),
+                detail=ERROR_MESSAGES.UNAUTHORIZED,
             )
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+
+############################
+# UpdateModelById
+############################
+
+
+@router.post("/model/update", response_model=Optional[ModelModel])
+async def update_model_by_id(
+    id: str,
+    form_data: ModelForm,
+    user=Depends(get_verified_user),
+):
+    model = Models.get_model_by_id(id)
+
+    if not model:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    model = Models.update_model_by_id(id, form_data)
+    return model
 
 
 ############################
@@ -98,7 +164,26 @@ async def update_model_by_id(
 ############################
 
 
-@router.delete("/delete", response_model=bool)
-async def delete_model_by_id(id: str, user=Depends(get_admin_user)):
+@router.delete("/model/delete", response_model=bool)
+async def delete_model_by_id(id: str, user=Depends(get_verified_user)):
+    model = Models.get_model_by_id(id)
+    if not model:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if model.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
     result = Models.delete_model_by_id(id)
     return result
+
+
+@router.delete("/delete/all", response_model=bool)
+async def delete_all_models(user=Depends(get_admin_user)):
+    result = Models.delete_all_models()
+    return result

backend/open_webui/apps/webui/routers/prompts.py

@@ -1,9 +1,15 @@
 from typing import Optional
 
-from open_webui.apps.webui.models.prompts import PromptForm, PromptModel, Prompts
+from open_webui.apps.webui.models.prompts import (
+    PromptForm,
+    PromptUserResponse,
+    PromptModel,
+    Prompts,
+)
 from open_webui.constants import ERROR_MESSAGES
-from fastapi import APIRouter, Depends, HTTPException, status
+from fastapi import APIRouter, Depends, HTTPException, status, Request
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access, has_permission
 
 router = APIRouter()
 
@@ -14,7 +20,22 @@ router = APIRouter()
 
 @router.get("/", response_model=list[PromptModel])
 async def get_prompts(user=Depends(get_verified_user)):
-    return Prompts.get_prompts()
+    if user.role == "admin":
+        prompts = Prompts.get_prompts()
+    else:
+        prompts = Prompts.get_prompts_by_user_id(user.id, "read")
+
+    return prompts
+
+
+@router.get("/list", response_model=list[PromptUserResponse])
+async def get_prompt_list(user=Depends(get_verified_user)):
+    if user.role == "admin":
+        prompts = Prompts.get_prompts()
+    else:
+        prompts = Prompts.get_prompts_by_user_id(user.id, "write")
+
+    return prompts
 
 
 ############################
@@ -23,7 +44,17 @@ async def get_prompts(user=Depends(get_verified_user)):
 
 
 @router.post("/create", response_model=Optional[PromptModel])
-async def create_new_prompt(form_data: PromptForm, user=Depends(get_admin_user)):
+async def create_new_prompt(
+    request: Request, form_data: PromptForm, user=Depends(get_verified_user)
+):
+    if user.role != "admin" and not has_permission(
+        user.id, "workspace.prompts", request.app.state.config.USER_PERMISSIONS
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
     prompt = Prompts.get_prompt_by_command(form_data.command)
     if prompt is None:
         prompt = Prompts.insert_new_prompt(user.id, form_data)
@@ -50,7 +81,12 @@ async def get_prompt_by_command(command: str, user=Depends(get_verified_user)):
     prompt = Prompts.get_prompt_by_command(f"/{command}")
 
     if prompt:
-        return prompt
+        if (
+            user.role == "admin"
+            or prompt.user_id == user.id
+            or has_access(user.id, "read", prompt.access_control)
+        ):
+            return prompt
     else:
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
@@ -67,8 +103,21 @@ async def get_prompt_by_command(command: str, user=Depends(get_verified_user)):
 async def update_prompt_by_command(
     command: str,
     form_data: PromptForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
+    prompt = Prompts.get_prompt_by_command(f"/{command}")
+    if not prompt:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if prompt.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     prompt = Prompts.update_prompt_by_command(f"/{command}", form_data)
     if prompt:
         return prompt
@@ -85,6 +134,19 @@ async def update_prompt_by_command(
 
 
 @router.delete("/command/{command}/delete", response_model=bool)
-async def delete_prompt_by_command(command: str, user=Depends(get_admin_user)):
+async def delete_prompt_by_command(command: str, user=Depends(get_verified_user)):
+    prompt = Prompts.get_prompt_by_command(f"/{command}")
+    if not prompt:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if prompt.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
     result = Prompts.delete_prompt_by_command(f"/{command}")
     return result

backend/open_webui/apps/webui/routers/tools.py

@@ -2,50 +2,82 @@ import os
 from pathlib import Path
 from typing import Optional
 
-from open_webui.apps.webui.models.tools import ToolForm, ToolModel, ToolResponse, Tools
-from open_webui.apps.webui.utils import load_toolkit_module_by_id, replace_imports
+from open_webui.apps.webui.models.tools import (
+    ToolForm,
+    ToolModel,
+    ToolResponse,
+    ToolUserResponse,
+    Tools,
+)
+from open_webui.apps.webui.utils import load_tools_module_by_id, replace_imports
 from open_webui.config import CACHE_DIR, DATA_DIR
 from open_webui.constants import ERROR_MESSAGES
 from fastapi import APIRouter, Depends, HTTPException, Request, status
 from open_webui.utils.tools import get_tools_specs
 from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.access_control import has_access, has_permission
 
 
 router = APIRouter()
 
 ############################
-# GetToolkits
+# GetTools
 ############################
 
 
-@router.get("/", response_model=list[ToolResponse])
-async def get_toolkits(user=Depends(get_verified_user)):
-    toolkits = [toolkit for toolkit in Tools.get_tools()]
-    return toolkits
+@router.get("/", response_model=list[ToolUserResponse])
+async def get_tools(user=Depends(get_verified_user)):
+    if user.role == "admin":
+        tools = Tools.get_tools()
+    else:
+        tools = Tools.get_tools_by_user_id(user.id, "read")
+    return tools
 
 
 ############################
-# ExportToolKits
+# GetToolList
+############################
+
+
+@router.get("/list", response_model=list[ToolUserResponse])
+async def get_tool_list(user=Depends(get_verified_user)):
+    if user.role == "admin":
+        tools = Tools.get_tools()
+    else:
+        tools = Tools.get_tools_by_user_id(user.id, "write")
+    return tools
+
+
+############################
+# ExportTools
 ############################
 
 
 @router.get("/export", response_model=list[ToolModel])
-async def get_toolkits(user=Depends(get_admin_user)):
-    toolkits = [toolkit for toolkit in Tools.get_tools()]
-    return toolkits
+async def export_tools(user=Depends(get_admin_user)):
+    tools = Tools.get_tools()
+    return tools
 
 
 ############################
-# CreateNewToolKit
+# CreateNewTools
 ############################
 
 
 @router.post("/create", response_model=Optional[ToolResponse])
-async def create_new_toolkit(
+async def create_new_tools(
     request: Request,
     form_data: ToolForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
+    if user.role != "admin" and not has_permission(
+        user.id, "workspace.knowledge", request.app.state.config.USER_PERMISSIONS
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
     if not form_data.id.isidentifier():
         raise HTTPException(
             status_code=status.HTTP_400_BAD_REQUEST,
@@ -54,30 +86,30 @@ async def create_new_toolkit(
 
     form_data.id = form_data.id.lower()
 
-    toolkit = Tools.get_tool_by_id(form_data.id)
-    if toolkit is None:
+    tools = Tools.get_tool_by_id(form_data.id)
+    if tools is None:
         try:
             form_data.content = replace_imports(form_data.content)
-            toolkit_module, frontmatter = load_toolkit_module_by_id(
+            tools_module, frontmatter = load_tools_module_by_id(
                 form_data.id, content=form_data.content
             )
             form_data.meta.manifest = frontmatter
 
             TOOLS = request.app.state.TOOLS
-            TOOLS[form_data.id] = toolkit_module
+            TOOLS[form_data.id] = tools_module
 
             specs = get_tools_specs(TOOLS[form_data.id])
-            toolkit = Tools.insert_new_tool(user.id, form_data, specs)
+            tools = Tools.insert_new_tool(user.id, form_data, specs)
 
             tool_cache_dir = Path(CACHE_DIR) / "tools" / form_data.id
             tool_cache_dir.mkdir(parents=True, exist_ok=True)
 
-            if toolkit:
-                return toolkit
+            if tools:
+                return tools
             else:
                 raise HTTPException(
                     status_code=status.HTTP_400_BAD_REQUEST,
-                    detail=ERROR_MESSAGES.DEFAULT("Error creating toolkit"),
+                    detail=ERROR_MESSAGES.DEFAULT("Error creating tools"),
                 )
         except Exception as e:
             print(e)
@@ -93,16 +125,21 @@ async def create_new_toolkit(
 
 
 ############################
-# GetToolkitById
+# GetToolsById
 ############################
 
 
 @router.get("/id/{id}", response_model=Optional[ToolModel])
-async def get_toolkit_by_id(id: str, user=Depends(get_admin_user)):
-    toolkit = Tools.get_tool_by_id(id)
-
-    if toolkit:
-        return toolkit
+async def get_tools_by_id(id: str, user=Depends(get_verified_user)):
+    tools = Tools.get_tool_by_id(id)
+
+    if tools:
+        if (
+            user.role == "admin"
+            or tools.user_id == user.id
+            or has_access(user.id, "read", tools.access_control)
+        ):
+            return tools
     else:
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
@@ -111,26 +148,39 @@ async def get_toolkit_by_id(id: str, user=Depends(get_admin_user)):
 
 
 ############################
-# UpdateToolkitById
+# UpdateToolsById
 ############################
 
 
 @router.post("/id/{id}/update", response_model=Optional[ToolModel])
-async def update_toolkit_by_id(
+async def update_tools_by_id(
     request: Request,
     id: str,
     form_data: ToolForm,
-    user=Depends(get_admin_user),
+    user=Depends(get_verified_user),
 ):
+    tools = Tools.get_tool_by_id(id)
+    if not tools:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
+
+    if tools.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
     try:
         form_data.content = replace_imports(form_data.content)
-        toolkit_module, frontmatter = load_toolkit_module_by_id(
+        tools_module, frontmatter = load_tools_module_by_id(
             id, content=form_data.content
         )
         form_data.meta.manifest = frontmatter
 
         TOOLS = request.app.state.TOOLS
-        TOOLS[id] = toolkit_module
+        TOOLS[id] = tools_module
 
         specs = get_tools_specs(TOOLS[id])
 
@@ -140,14 +190,14 @@ async def update_toolkit_by_id(
         }
 
         print(updated)
-        toolkit = Tools.update_tool_by_id(id, updated)
+        tools = Tools.update_tool_by_id(id, updated)
 
-        if toolkit:
-            return toolkit
+        if tools:
+            return tools
         else:
             raise HTTPException(
                 status_code=status.HTTP_400_BAD_REQUEST,
-                detail=ERROR_MESSAGES.DEFAULT("Error updating toolkit"),
+                detail=ERROR_MESSAGES.DEFAULT("Error updating tools"),
             )
 
     except Exception as e:
@@ -158,14 +208,28 @@ async def update_toolkit_by_id(
 
 
 ############################
-# DeleteToolkitById
+# DeleteToolsById
 ############################
 
 
 @router.delete("/id/{id}/delete", response_model=bool)
-async def delete_toolkit_by_id(request: Request, id: str, user=Depends(get_admin_user)):
-    result = Tools.delete_tool_by_id(id)
+async def delete_tools_by_id(
+    request: Request, id: str, user=Depends(get_verified_user)
+):
+    tools = Tools.get_tool_by_id(id)
+    if not tools:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.NOT_FOUND,
+        )
 
+    if tools.user_id != user.id and user.role != "admin":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.UNAUTHORIZED,
+        )
+
+    result = Tools.delete_tool_by_id(id)
     if result:
         TOOLS = request.app.state.TOOLS
         if id in TOOLS:
@@ -180,9 +244,9 @@ async def delete_toolkit_by_id(request: Request, id: str, user=Depends(get_admin
 
 
 @router.get("/id/{id}/valves", response_model=Optional[dict])
-async def get_toolkit_valves_by_id(id: str, user=Depends(get_admin_user)):
-    toolkit = Tools.get_tool_by_id(id)
-    if toolkit:
+async def get_tools_valves_by_id(id: str, user=Depends(get_verified_user)):
+    tools = Tools.get_tool_by_id(id)
+    if tools:
         try:
             valves = Tools.get_tool_valves_by_id(id)
             return valves
@@ -204,19 +268,19 @@ async def get_toolkit_valves_by_id(id: str, user=Depends(get_admin_user)):
 
 
 @router.get("/id/{id}/valves/spec", response_model=Optional[dict])
-async def get_toolkit_valves_spec_by_id(
-    request: Request, id: str, user=Depends(get_admin_user)
+async def get_tools_valves_spec_by_id(
+    request: Request, id: str, user=Depends(get_verified_user)
 ):
-    toolkit = Tools.get_tool_by_id(id)
-    if toolkit:
+    tools = Tools.get_tool_by_id(id)
+    if tools:
         if id in request.app.state.TOOLS:
-            toolkit_module = request.app.state.TOOLS[id]
+            tools_module = request.app.state.TOOLS[id]
         else:
-            toolkit_module, _ = load_toolkit_module_by_id(id)
-            request.app.state.TOOLS[id] = toolkit_module
+            tools_module, _ = load_tools_module_by_id(id)
+            request.app.state.TOOLS[id] = tools_module
 
-        if hasattr(toolkit_module, "Valves"):
-            Valves = toolkit_module.Valves
+        if hasattr(tools_module, "Valves"):
+            Valves = tools_module.Valves
             return Valves.schema()
         return None
     else:
@@ -232,19 +296,19 @@ async def get_toolkit_valves_spec_by_id(
 
 
 @router.post("/id/{id}/valves/update", response_model=Optional[dict])
-async def update_toolkit_valves_by_id(
-    request: Request, id: str, form_data: dict, user=Depends(get_admin_user)
+async def update_tools_valves_by_id(
+    request: Request, id: str, form_data: dict, user=Depends(get_verified_user)
 ):
-    toolkit = Tools.get_tool_by_id(id)
-    if toolkit:
+    tools = Tools.get_tool_by_id(id)
+    if tools:
         if id in request.app.state.TOOLS:
-            toolkit_module = request.app.state.TOOLS[id]
+            tools_module = request.app.state.TOOLS[id]
         else:
-            toolkit_module, _ = load_toolkit_module_by_id(id)
-            request.app.state.TOOLS[id] = toolkit_module
+            tools_module, _ = load_tools_module_by_id(id)
+            request.app.state.TOOLS[id] = tools_module
 
-        if hasattr(toolkit_module, "Valves"):
-            Valves = toolkit_module.Valves
+        if hasattr(tools_module, "Valves"):
+            Valves = tools_module.Valves
 
             try:
                 form_data = {k: v for k, v in form_data.items() if v is not None}
@@ -276,9 +340,9 @@ async def update_toolkit_valves_by_id(
 
 
 @router.get("/id/{id}/valves/user", response_model=Optional[dict])
-async def get_toolkit_user_valves_by_id(id: str, user=Depends(get_verified_user)):
-    toolkit = Tools.get_tool_by_id(id)
-    if toolkit:
+async def get_tools_user_valves_by_id(id: str, user=Depends(get_verified_user)):
+    tools = Tools.get_tool_by_id(id)
+    if tools:
         try:
             user_valves = Tools.get_user_valves_by_id_and_user_id(id, user.id)
             return user_valves
@@ -295,19 +359,19 @@ async def get_toolkit_user_valves_by_id(id: str, user=Depends(get_verified_user)
 
 
 @router.get("/id/{id}/valves/user/spec", response_model=Optional[dict])
-async def get_toolkit_user_valves_spec_by_id(
+async def get_tools_user_valves_spec_by_id(
     request: Request, id: str, user=Depends(get_verified_user)
 ):
-    toolkit = Tools.get_tool_by_id(id)
-    if toolkit:
+    tools = Tools.get_tool_by_id(id)
+    if tools:
         if id in request.app.state.TOOLS:
-            toolkit_module = request.app.state.TOOLS[id]
+            tools_module = request.app.state.TOOLS[id]
         else:
-            toolkit_module, _ = load_toolkit_module_by_id(id)
-            request.app.state.TOOLS[id] = toolkit_module
+            tools_module, _ = load_tools_module_by_id(id)
+            request.app.state.TOOLS[id] = tools_module
 
-        if hasattr(toolkit_module, "UserValves"):
-            UserValves = toolkit_module.UserValves
+        if hasattr(tools_module, "UserValves"):
+            UserValves = tools_module.UserValves
             return UserValves.schema()
         return None
     else:
@@ -318,20 +382,20 @@ async def get_toolkit_user_valves_spec_by_id(
 
 
 @router.post("/id/{id}/valves/user/update", response_model=Optional[dict])
-async def update_toolkit_user_valves_by_id(
+async def update_tools_user_valves_by_id(
     request: Request, id: str, form_data: dict, user=Depends(get_verified_user)
 ):
-    toolkit = Tools.get_tool_by_id(id)
+    tools = Tools.get_tool_by_id(id)
 
-    if toolkit:
+    if tools:
         if id in request.app.state.TOOLS:
-            toolkit_module = request.app.state.TOOLS[id]
+            tools_module = request.app.state.TOOLS[id]
         else:
-            toolkit_module, _ = load_toolkit_module_by_id(id)
-            request.app.state.TOOLS[id] = toolkit_module
+            tools_module, _ = load_tools_module_by_id(id)
+            request.app.state.TOOLS[id] = tools_module
 
-        if hasattr(toolkit_module, "UserValves"):
-            UserValves = toolkit_module.UserValves
+        if hasattr(tools_module, "UserValves"):
+            UserValves = tools_module.UserValves
 
             try:
                 form_data = {k: v for k, v in form_data.items() if v is not None}

backend/open_webui/apps/webui/routers/users.py

@@ -31,21 +31,58 @@ async def get_users(skip: int = 0, limit: int = 50, user=Depends(get_admin_user)
     return Users.get_users(skip, limit)
 
 
+############################
+# User Groups
+############################
+
+
+@router.get("/groups")
+async def get_user_groups(user=Depends(get_verified_user)):
+    return Users.get_user_groups(user.id)
+
+
 ############################
 # User Permissions
 ############################
 
 
-@router.get("/permissions/user")
+@router.get("/permissions")
+async def get_user_permissisions(user=Depends(get_verified_user)):
+    return Users.get_user_groups(user.id)
+
+
+############################
+# User Default Permissions
+############################
+class WorkspacePermissions(BaseModel):
+    models: bool
+    knowledge: bool
+    prompts: bool
+    tools: bool
+
+
+class ChatPermissions(BaseModel):
+    file_upload: bool
+    delete: bool
+    edit: bool
+    temporary: bool
+
+
+class UserPermissions(BaseModel):
+    workspace: WorkspacePermissions
+    chat: ChatPermissions
+
+
+@router.get("/default/permissions")
 async def get_user_permissions(request: Request, user=Depends(get_admin_user)):
     return request.app.state.config.USER_PERMISSIONS
 
 
-@router.post("/permissions/user")
+@router.post("/default/permissions")
 async def update_user_permissions(
-    request: Request, form_data: dict, user=Depends(get_admin_user)
+    request: Request, form_data: UserPermissions, user=Depends(get_admin_user)
 ):
-    request.app.state.config.USER_PERMISSIONS = form_data
+    request.app.state.config.USER_PERMISSIONS = form_data.model_dump()
     return request.app.state.config.USER_PERMISSIONS
 
 

backend/open_webui/apps/webui/utils.py

@@ -63,7 +63,7 @@ def replace_imports(content):
     return content
 
 
-def load_toolkit_module_by_id(toolkit_id, content=None):
+def load_tools_module_by_id(toolkit_id, content=None):
 
     if content is None:
         tool = Tools.get_tool_by_id(toolkit_id)

backend/open_webui/config.py

@@ -20,6 +20,7 @@ from open_webui.env import (
     WEBUI_FAVICON_URL,
     WEBUI_NAME,
     log,
+    DATABASE_URL,
 )
 from pydantic import BaseModel
 from sqlalchemy import JSON, Column, DateTime, Integer, func
@@ -264,6 +265,13 @@ class AppConfig:
 # WEBUI_AUTH (Required for security)
 ####################################
 
+ENABLE_API_KEY = PersistentConfig(
+    "ENABLE_API_KEY",
+    "auth.api_key.enable",
+    os.environ.get("ENABLE_API_KEY", "True").lower() == "true",
+)
+
+
 JWT_EXPIRES_IN = PersistentConfig(
     "JWT_EXPIRES_IN", "auth.jwt_expiry", os.environ.get("JWT_EXPIRES_IN", "-1")
 )
@@ -606,6 +614,12 @@ OLLAMA_BASE_URLS = PersistentConfig(
     "OLLAMA_BASE_URLS", "ollama.base_urls", OLLAMA_BASE_URLS
 )
 
+OLLAMA_API_CONFIGS = PersistentConfig(
+    "OLLAMA_API_CONFIGS",
+    "ollama.api_configs",
+    {},
+)
+
 ####################################
 # OPENAI_API
 ####################################
@@ -646,15 +660,20 @@ OPENAI_API_BASE_URLS = PersistentConfig(
     "OPENAI_API_BASE_URLS", "openai.api_base_urls", OPENAI_API_BASE_URLS
 )
 
-OPENAI_API_KEY = ""
+OPENAI_API_CONFIGS = PersistentConfig(
+    "OPENAI_API_CONFIGS",
+    "openai.api_configs",
+    {},
+)
 
+# Get the actual OpenAI API key based on the base URL
+OPENAI_API_KEY = ""
 try:
     OPENAI_API_KEY = OPENAI_API_KEYS.value[
         OPENAI_API_BASE_URLS.value.index("https://api.openai.com/v1")
     ]
 except Exception:
     pass
-
 OPENAI_API_BASE_URL = "https://api.openai.com/v1"
 
 ####################################
@@ -727,12 +746,36 @@ DEFAULT_USER_ROLE = PersistentConfig(
     os.getenv("DEFAULT_USER_ROLE", "pending"),
 )
 
-USER_PERMISSIONS_CHAT_DELETION = (
-    os.environ.get("USER_PERMISSIONS_CHAT_DELETION", "True").lower() == "true"
+
+USER_PERMISSIONS_WORKSPACE_MODELS_ACCESS = (
+    os.environ.get("USER_PERMISSIONS_WORKSPACE_MODELS_ACCESS", "False").lower()
+    == "true"
+)
+
+USER_PERMISSIONS_WORKSPACE_KNOWLEDGE_ACCESS = (
+    os.environ.get("USER_PERMISSIONS_WORKSPACE_KNOWLEDGE_ACCESS", "False").lower()
+    == "true"
+)
+
+USER_PERMISSIONS_WORKSPACE_PROMPTS_ACCESS = (
+    os.environ.get("USER_PERMISSIONS_WORKSPACE_PROMPTS_ACCESS", "False").lower()
+    == "true"
+)
+
+USER_PERMISSIONS_WORKSPACE_TOOLS_ACCESS = (
+    os.environ.get("USER_PERMISSIONS_WORKSPACE_TOOLS_ACCESS", "False").lower() == "true"
+)
+
+USER_PERMISSIONS_CHAT_FILE_UPLOAD = (
+    os.environ.get("USER_PERMISSIONS_CHAT_FILE_UPLOAD", "True").lower() == "true"
+)
+
+USER_PERMISSIONS_CHAT_DELETE = (
+    os.environ.get("USER_PERMISSIONS_CHAT_DELETE", "True").lower() == "true"
 )
 
-USER_PERMISSIONS_CHAT_EDITING = (
-    os.environ.get("USER_PERMISSIONS_CHAT_EDITING", "True").lower() == "true"
+USER_PERMISSIONS_CHAT_EDIT = (
+    os.environ.get("USER_PERMISSIONS_CHAT_EDIT", "True").lower() == "true"
 )
 
 USER_PERMISSIONS_CHAT_TEMPORARY = (
@@ -741,13 +784,20 @@ USER_PERMISSIONS_CHAT_TEMPORARY = (
 
 USER_PERMISSIONS = PersistentConfig(
     "USER_PERMISSIONS",
-    "ui.user_permissions",
+    "user.permissions",
     {
+        "workspace": {
+            "models": USER_PERMISSIONS_WORKSPACE_MODELS_ACCESS,
+            "knowledge": USER_PERMISSIONS_WORKSPACE_KNOWLEDGE_ACCESS,
+            "prompts": USER_PERMISSIONS_WORKSPACE_PROMPTS_ACCESS,
+            "tools": USER_PERMISSIONS_WORKSPACE_TOOLS_ACCESS,
+        },
         "chat": {
-            "deletion": USER_PERMISSIONS_CHAT_DELETION,
-            "editing": USER_PERMISSIONS_CHAT_EDITING,
+            "file_upload": USER_PERMISSIONS_CHAT_FILE_UPLOAD,
+            "delete": USER_PERMISSIONS_CHAT_DELETE,
+            "edit": USER_PERMISSIONS_CHAT_EDIT,
             "temporary": USER_PERMISSIONS_CHAT_TEMPORARY,
-        }
+        },
     },
 )
 
@@ -773,18 +823,6 @@ DEFAULT_ARENA_MODEL = {
     },
 }
 
-ENABLE_MODEL_FILTER = PersistentConfig(
-    "ENABLE_MODEL_FILTER",
-    "model_filter.enable",
-    os.environ.get("ENABLE_MODEL_FILTER", "False").lower() == "true",
-)
-MODEL_FILTER_LIST = os.environ.get("MODEL_FILTER_LIST", "")
-MODEL_FILTER_LIST = PersistentConfig(
-    "MODEL_FILTER_LIST",
-    "model_filter.list",
-    [model.strip() for model in MODEL_FILTER_LIST.split(";")],
-)
-
 WEBHOOK_URL = PersistentConfig(
     "WEBHOOK_URL", "webhook_url", os.environ.get("WEBHOOK_URL", "")
 )
@@ -904,19 +942,55 @@ TAGS_GENERATION_PROMPT_TEMPLATE = PersistentConfig(
     os.environ.get("TAGS_GENERATION_PROMPT_TEMPLATE", ""),
 )
 
-ENABLE_SEARCH_QUERY = PersistentConfig(
-    "ENABLE_SEARCH_QUERY",
-    "task.search.enable",
-    os.environ.get("ENABLE_SEARCH_QUERY", "True").lower() == "true",
+ENABLE_TAGS_GENERATION = PersistentConfig(
+    "ENABLE_TAGS_GENERATION",
+    "task.tags.enable",
+    os.environ.get("ENABLE_TAGS_GENERATION", "True").lower() == "true",
+)
+
+
+ENABLE_SEARCH_QUERY_GENERATION = PersistentConfig(
+    "ENABLE_SEARCH_QUERY_GENERATION",
+    "task.query.search.enable",
+    os.environ.get("ENABLE_SEARCH_QUERY_GENERATION", "True").lower() == "true",
+)
+
+ENABLE_RETRIEVAL_QUERY_GENERATION = PersistentConfig(
+    "ENABLE_RETRIEVAL_QUERY_GENERATION",
+    "task.query.retrieval.enable",
+    os.environ.get("ENABLE_RETRIEVAL_QUERY_GENERATION", "True").lower() == "true",
 )
 
 
-SEARCH_QUERY_GENERATION_PROMPT_TEMPLATE = PersistentConfig(
-    "SEARCH_QUERY_GENERATION_PROMPT_TEMPLATE",
-    "task.search.prompt_template",
-    os.environ.get("SEARCH_QUERY_GENERATION_PROMPT_TEMPLATE", ""),
+QUERY_GENERATION_PROMPT_TEMPLATE = PersistentConfig(
+    "QUERY_GENERATION_PROMPT_TEMPLATE",
+    "task.query.prompt_template",
+    os.environ.get("QUERY_GENERATION_PROMPT_TEMPLATE", ""),
 )
 
+DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE = """### Task:
+Based on the chat history, determine whether a search is necessary, and if so, generate a 1-3 broad search queries to retrieve comprehensive and updated information. If no search is required, return an empty list.
+
+### Guidelines:
+- Respond exclusively with a JSON object.
+- If a search query is needed, return an object like: { "queries": ["query1", "query2"] } where each query is distinct and concise.
+- If no search query is necessary, output should be: { "queries": [] }
+- Default to suggesting a search query to ensure accurate and updated information, unless it is definitively clear no search is required.
+- Be concise, focusing strictly on composing search queries with no additional commentary or text.
+- When in doubt, prefer to suggest a search for comprehensiveness.
+- Today's date is: {{CURRENT_DATE}}
+
+### Output:
+JSON format: {
+  "queries": ["query1", "query2"]
+}
+
+### Chat History:
+<chat_history>
+{{MESSAGES:END:6}}
+</chat_history>
+"""
+
 
 TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = PersistentConfig(
     "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE",
@@ -937,6 +1011,8 @@ CHROMA_TENANT = os.environ.get("CHROMA_TENANT", chromadb.DEFAULT_TENANT)
 CHROMA_DATABASE = os.environ.get("CHROMA_DATABASE", chromadb.DEFAULT_DATABASE)
 CHROMA_HTTP_HOST = os.environ.get("CHROMA_HTTP_HOST", "")
 CHROMA_HTTP_PORT = int(os.environ.get("CHROMA_HTTP_PORT", "8000"))
+CHROMA_CLIENT_AUTH_PROVIDER = os.environ.get("CHROMA_CLIENT_AUTH_PROVIDER", "")
+CHROMA_CLIENT_AUTH_CREDENTIALS = os.environ.get("CHROMA_CLIENT_AUTH_CREDENTIALS", "")
 # Comma-separated list of header=value pairs
 CHROMA_HTTP_HEADERS = os.environ.get("CHROMA_HTTP_HEADERS", "")
 if CHROMA_HTTP_HEADERS:
@@ -954,6 +1030,21 @@ MILVUS_URI = os.environ.get("MILVUS_URI", f"{DATA_DIR}/vector_db/milvus.db")
 
 # Qdrant
 QDRANT_URI = os.environ.get("QDRANT_URI", None)
+QDRANT_API_KEY = os.environ.get("QDRANT_API_KEY", None)
+
+# OpenSearch
+OPENSEARCH_URI = os.environ.get("OPENSEARCH_URI", "https://localhost:9200")
+OPENSEARCH_SSL = os.environ.get("OPENSEARCH_SSL", True)
+OPENSEARCH_CERT_VERIFY = os.environ.get("OPENSEARCH_CERT_VERIFY", False)
+OPENSEARCH_USERNAME = os.environ.get("OPENSEARCH_USERNAME", None)
+OPENSEARCH_PASSWORD = os.environ.get("OPENSEARCH_PASSWORD", None)
+
+# Pgvector
+PGVECTOR_DB_URL = os.environ.get("PGVECTOR_DB_URL", DATABASE_URL)
+if VECTOR_DB == "pgvector" and not PGVECTOR_DB_URL.startswith("postgres"):
+    raise ValueError(
+        "Pgvector requires setting PGVECTOR_DB_URL or using Postgres with vector extension as the primary database."
+    )
 
 ####################################
 # Information Retrieval (RAG)
@@ -1033,11 +1124,11 @@ RAG_EMBEDDING_MODEL = PersistentConfig(
 log.info(f"Embedding model set: {RAG_EMBEDDING_MODEL.value}")
 
 RAG_EMBEDDING_MODEL_AUTO_UPDATE = (
-    os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "").lower() == "true"
+    os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "True").lower() == "true"
 )
 
 RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE = (
-    os.environ.get("RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE", "").lower() == "true"
+    os.environ.get("RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE", "True").lower() == "true"
 )
 
 RAG_EMBEDDING_BATCH_SIZE = PersistentConfig(
@@ -1058,11 +1149,11 @@ if RAG_RERANKING_MODEL.value != "":
     log.info(f"Reranking model set: {RAG_RERANKING_MODEL.value}")
 
 RAG_RERANKING_MODEL_AUTO_UPDATE = (
-    os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "").lower() == "true"
+    os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "True").lower() == "true"
 )
 
 RAG_RERANKING_MODEL_TRUST_REMOTE_CODE = (
-    os.environ.get("RAG_RERANKING_MODEL_TRUST_REMOTE_CODE", "").lower() == "true"
+    os.environ.get("RAG_RERANKING_MODEL_TRUST_REMOTE_CODE", "True").lower() == "true"
 )
 
 
@@ -1127,6 +1218,19 @@ RAG_OPENAI_API_KEY = PersistentConfig(
     os.getenv("RAG_OPENAI_API_KEY", OPENAI_API_KEY),
 )
 
+RAG_OLLAMA_BASE_URL = PersistentConfig(
+    "RAG_OLLAMA_BASE_URL",
+    "rag.ollama.url",
+    os.getenv("RAG_OLLAMA_BASE_URL", OLLAMA_BASE_URL),
+)
+
+RAG_OLLAMA_API_KEY = PersistentConfig(
+    "RAG_OLLAMA_API_KEY",
+    "rag.ollama.key",
+    os.getenv("RAG_OLLAMA_API_KEY", ""),
+)
+
+
 ENABLE_RAG_LOCAL_WEB_FETCH = (
     os.getenv("ENABLE_RAG_LOCAL_WEB_FETCH", "False").lower() == "true"
 )
@@ -1222,6 +1326,12 @@ TAVILY_API_KEY = PersistentConfig(
     os.getenv("TAVILY_API_KEY", ""),
 )
 
+JINA_API_KEY = PersistentConfig(
+    "JINA_API_KEY",
+    "rag.web.search.jina_api_key",
+    os.getenv("JINA_API_KEY", ""),
+)
+
 SEARCHAPI_API_KEY = PersistentConfig(
     "SEARCHAPI_API_KEY",
     "rag.web.search.searchapi_api_key",
@@ -1234,6 +1344,21 @@ SEARCHAPI_ENGINE = PersistentConfig(
     os.getenv("SEARCHAPI_ENGINE", ""),
 )
 
+BING_SEARCH_V7_ENDPOINT = PersistentConfig(
+    "BING_SEARCH_V7_ENDPOINT",
+    "rag.web.search.bing_search_v7_endpoint",
+    os.environ.get(
+        "BING_SEARCH_V7_ENDPOINT", "https://api.bing.microsoft.com/v7.0/search"
+    ),
+)
+
+BING_SEARCH_V7_SUBSCRIPTION_KEY = PersistentConfig(
+    "BING_SEARCH_V7_SUBSCRIPTION_KEY",
+    "rag.web.search.bing_search_v7_subscription_key",
+    os.environ.get("BING_SEARCH_V7_SUBSCRIPTION_KEY", ""),
+)
+
+
 RAG_WEB_SEARCH_RESULT_COUNT = PersistentConfig(
     "RAG_WEB_SEARCH_RESULT_COUNT",
     "rag.web.search.result_count",
@@ -1285,7 +1410,7 @@ AUTOMATIC1111_CFG_SCALE = PersistentConfig(
 
 
 AUTOMATIC1111_SAMPLER = PersistentConfig(
-    "AUTOMATIC1111_SAMPLERE",
+    "AUTOMATIC1111_SAMPLER",
     "image_generation.automatic1111.sampler",
     (
         os.environ.get("AUTOMATIC1111_SAMPLER")
@@ -1554,3 +1679,74 @@ AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT = PersistentConfig(
         "AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT", "audio-24khz-160kbitrate-mono-mp3"
     ),
 )
+
+
+####################################
+# LDAP
+####################################
+
+ENABLE_LDAP = PersistentConfig(
+    "ENABLE_LDAP",
+    "ldap.enable",
+    os.environ.get("ENABLE_LDAP", "false").lower() == "true",
+)
+
+LDAP_SERVER_LABEL = PersistentConfig(
+    "LDAP_SERVER_LABEL",
+    "ldap.server.label",
+    os.environ.get("LDAP_SERVER_LABEL", "LDAP Server"),
+)
+
+LDAP_SERVER_HOST = PersistentConfig(
+    "LDAP_SERVER_HOST",
+    "ldap.server.host",
+    os.environ.get("LDAP_SERVER_HOST", "localhost"),
+)
+
+LDAP_SERVER_PORT = PersistentConfig(
+    "LDAP_SERVER_PORT",
+    "ldap.server.port",
+    int(os.environ.get("LDAP_SERVER_PORT", "389")),
+)
+
+LDAP_ATTRIBUTE_FOR_USERNAME = PersistentConfig(
+    "LDAP_ATTRIBUTE_FOR_USERNAME",
+    "ldap.server.attribute_for_username",
+    os.environ.get("LDAP_ATTRIBUTE_FOR_USERNAME", "uid"),
+)
+
+LDAP_APP_DN = PersistentConfig(
+    "LDAP_APP_DN", "ldap.server.app_dn", os.environ.get("LDAP_APP_DN", "")
+)
+
+LDAP_APP_PASSWORD = PersistentConfig(
+    "LDAP_APP_PASSWORD",
+    "ldap.server.app_password",
+    os.environ.get("LDAP_APP_PASSWORD", ""),
+)
+
+LDAP_SEARCH_BASE = PersistentConfig(
+    "LDAP_SEARCH_BASE", "ldap.server.users_dn", os.environ.get("LDAP_SEARCH_BASE", "")
+)
+
+LDAP_SEARCH_FILTERS = PersistentConfig(
+    "LDAP_SEARCH_FILTER",
+    "ldap.server.search_filter",
+    os.environ.get("LDAP_SEARCH_FILTER", ""),
+)
+
+LDAP_USE_TLS = PersistentConfig(
+    "LDAP_USE_TLS",
+    "ldap.server.use_tls",
+    os.environ.get("LDAP_USE_TLS", "True").lower() == "true",
+)
+
+LDAP_CA_CERT_FILE = PersistentConfig(
+    "LDAP_CA_CERT_FILE",
+    "ldap.server.ca_cert_file",
+    os.environ.get("LDAP_CA_CERT_FILE", ""),
+)
+
+LDAP_CIPHERS = PersistentConfig(
+    "LDAP_CIPHERS", "ldap.server.ciphers", os.environ.get("LDAP_CIPHERS", "ALL")
+)

backend/open_webui/constants.py

@@ -62,6 +62,7 @@ class ERROR_MESSAGES(str, Enum):
     NOT_FOUND = "We could not find what you're looking for :/"
     USER_NOT_FOUND = "We could not find what you're looking for :/"
     API_KEY_NOT_FOUND = "Oops! It looks like there's a hiccup. The API key is missing. Please make sure to provide a valid API key to access this feature."
+    API_KEY_NOT_ALLOWED = "Use of API key is not enabled in the environment."
 
     MALICIOUS = "Unusual activities detected, please try again in a few minutes."
 
@@ -75,6 +76,7 @@ class ERROR_MESSAGES(str, Enum):
     OPENAI_NOT_FOUND = lambda name="": "OpenAI API was not found"
     OLLAMA_NOT_FOUND = "WebUI could not connect to Ollama"
     CREATE_API_KEY_ERROR = "Oops! Something went wrong while creating your API key. Please try again later. If the issue persists, contact support for assistance."
+    API_KEY_CREATION_NOT_ALLOWED = "API key creation is not allowed in the environment."
 
     EMPTY_CONTENT = "The content provided is empty. Please ensure that there is text or data present before proceeding."
 

backend/open_webui/env.py

@@ -195,6 +195,15 @@ CHANGELOG = changelog_json
 
 SAFE_MODE = os.environ.get("SAFE_MODE", "false").lower() == "true"
 
+####################################
+# ENABLE_FORWARD_USER_INFO_HEADERS
+####################################
+
+ENABLE_FORWARD_USER_INFO_HEADERS = (
+    os.environ.get("ENABLE_FORWARD_USER_INFO_HEADERS", "False").lower() == "true"
+)
+
+
 ####################################
 # WEBUI_BUILD_HASH
 ####################################

backend/open_webui/migrations/versions/922e7a387820_add_group_table.py

@@ -0,0 +1,85 @@
+"""Add group table
+
+Revision ID: 922e7a387820
+Revises: 4ace53fd72c8
+Create Date: 2024-11-14 03:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+revision = "922e7a387820"
+down_revision = "4ace53fd72c8"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    op.create_table(
+        "group",
+        sa.Column("id", sa.Text(), nullable=False, primary_key=True, unique=True),
+        sa.Column("user_id", sa.Text(), nullable=True),
+        sa.Column("name", sa.Text(), nullable=True),
+        sa.Column("description", sa.Text(), nullable=True),
+        sa.Column("data", sa.JSON(), nullable=True),
+        sa.Column("meta", sa.JSON(), nullable=True),
+        sa.Column("permissions", sa.JSON(), nullable=True),
+        sa.Column("user_ids", sa.JSON(), nullable=True),
+        sa.Column("created_at", sa.BigInteger(), nullable=True),
+        sa.Column("updated_at", sa.BigInteger(), nullable=True),
+    )
+
+    # Add 'access_control' column to 'model' table
+    op.add_column(
+        "model",
+        sa.Column("access_control", sa.JSON(), nullable=True),
+    )
+
+    # Add 'is_active' column to 'model' table
+    op.add_column(
+        "model",
+        sa.Column(
+            "is_active",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.sql.expression.true(),
+        ),
+    )
+
+    # Add 'access_control' column to 'knowledge' table
+    op.add_column(
+        "knowledge",
+        sa.Column("access_control", sa.JSON(), nullable=True),
+    )
+
+    # Add 'access_control' column to 'prompt' table
+    op.add_column(
+        "prompt",
+        sa.Column("access_control", sa.JSON(), nullable=True),
+    )
+
+    # Add 'access_control' column to 'tools' table
+    op.add_column(
+        "tool",
+        sa.Column("access_control", sa.JSON(), nullable=True),
+    )
+
+
+def downgrade():
+    op.drop_table("group")
+
+    # Drop 'access_control' column from 'model' table
+    op.drop_column("model", "access_control")
+
+    # Drop 'is_active' column from 'model' table
+    op.drop_column("model", "is_active")
+
+    # Drop 'access_control' column from 'knowledge' table
+    op.drop_column("knowledge", "access_control")
+
+    # Drop 'access_control' column from 'prompt' table
+    op.drop_column("prompt", "access_control")
+
+    # Drop 'access_control' column from 'tools' table
+    op.drop_column("tool", "access_control")

backend/open_webui/storage/provider.py

@@ -51,7 +51,10 @@ class StorageProvider:
 
         try:
             self.s3_client.upload_file(file_path, self.bucket_name, filename)
-            return open(file_path, "rb").read(), file_path
+            return (
+                open(file_path, "rb").read(),
+                "s3://" + self.bucket_name + "/" + filename,
+            )
         except ClientError as e:
             raise RuntimeError(f"Error uploading file to S3: {e}")
 

backend/open_webui/utils/access_control.py

@@ -0,0 +1,95 @@
+from typing import Optional, Union, List, Dict, Any
+from open_webui.apps.webui.models.groups import Groups
+import json
+
+
+def get_permissions(
+    user_id: str,
+    default_permissions: Dict[str, Any],
+) -> Dict[str, Any]:
+    """
+    Get all permissions for a user by combining the permissions of all groups the user is a member of.
+    If a permission is defined in multiple groups, the most permissive value is used (True > False).
+    Permissions are nested in a dict with the permission key as the key and a boolean as the value.
+    """
+
+    def combine_permissions(
+        permissions: Dict[str, Any], group_permissions: Dict[str, Any]
+    ) -> Dict[str, Any]:
+        """Combine permissions from multiple groups by taking the most permissive value."""
+        for key, value in group_permissions.items():
+            if isinstance(value, dict):
+                if key not in permissions:
+                    permissions[key] = {}
+                permissions[key] = combine_permissions(permissions[key], value)
+            else:
+                if key not in permissions:
+                    permissions[key] = value
+                else:
+                    permissions[key] = permissions[key] or value
+        return permissions
+
+    user_groups = Groups.get_groups_by_member_id(user_id)
+
+    # deep copy default permissions to avoid modifying the original dict
+    permissions = json.loads(json.dumps(default_permissions))
+
+    for group in user_groups:
+        group_permissions = group.permissions
+        permissions = combine_permissions(permissions, group_permissions)
+
+    return permissions
+
+
+def has_permission(
+    user_id: str,
+    permission_key: str,
+    default_permissions: Dict[str, bool] = {},
+) -> bool:
+    """
+    Check if a user has a specific permission by checking the group permissions
+    and falls back to default permissions if not found in any group.
+
+    Permission keys can be hierarchical and separated by dots ('.').
+    """
+
+    def get_permission(permissions: Dict[str, bool], keys: List[str]) -> bool:
+        """Traverse permissions dict using a list of keys (from dot-split permission_key)."""
+        for key in keys:
+            if key not in permissions:
+                return False  # If any part of the hierarchy is missing, deny access
+            permissions = permissions[key]  # Go one level deeper
+
+        return bool(permissions)  # Return the boolean at the final level
+
+    permission_hierarchy = permission_key.split(".")
+
+    # Retrieve user group permissions
+    user_groups = Groups.get_groups_by_member_id(user_id)
+
+    for group in user_groups:
+        group_permissions = group.permissions
+        if get_permission(group_permissions, permission_hierarchy):
+            return True
+
+    # Check default permissions afterwards if the group permissions don't allow it
+    return get_permission(default_permissions, permission_hierarchy)
+
+
+def has_access(
+    user_id: str,
+    type: str = "write",
+    access_control: Optional[dict] = None,
+) -> bool:
+    if access_control is None:
+        return type == "read"
+
+    user_groups = Groups.get_groups_by_member_id(user_id)
+    user_group_ids = [group.id for group in user_groups]
+    permission_access = access_control.get(type, {})
+    permitted_group_ids = permission_access.get("group_ids", [])
+    permitted_user_ids = permission_access.get("user_ids", [])
+
+    return user_id in permitted_user_ids or any(
+        group_id in permitted_group_ids for group_id in user_group_ids
+    )

backend/open_webui/utils/pdf_generator.py

@@ -54,18 +54,18 @@ class PDFGenerator:
         html_content = markdown(content, extensions=["pymdownx.extra"])
 
         html_message = f"""
-              <div class="message">
-                  <small> {date_str} </small>
-                  <div>
-                      <h2>
-                          <strong>{role.title()}</strong>
-                          <small class="text-muted">{model}</small>
-                      </h2>
-                  </div>
-                  <div class="markdown-section">
-                      {html_content}
-                  </div>
-              </div>
+            <div> {date_str} </div>
+            <div class="message">
+                <div>
+                    <h2>
+                        <strong>{role.title()}</strong>
+                        <span style="font-size: 12px; color: #888;">{model}</span>
+                    </h2>
+                </div>
+                <pre class="markdown-section">
+                    {content}
+                </pre>
+            </div>
           """
         return html_message
 

backend/open_webui/utils/security_headers.py

@@ -20,6 +20,7 @@ def set_security_headers() -> Dict[str, str]:
     This function reads specific environment variables and uses their values
     to set corresponding security headers. The headers that can be set are:
     - cache-control
+    - permissions-policy
     - strict-transport-security
     - referrer-policy
     - x-content-type-options
@@ -38,6 +39,7 @@ def set_security_headers() -> Dict[str, str]:
     header_setters = {
         "CACHE_CONTROL": set_cache_control,
         "HSTS": set_hsts,
+        "PERMISSIONS_POLICY": set_permissions_policy,
         "REFERRER_POLICY": set_referrer,
         "XCONTENT_TYPE": set_xcontent_type,
         "XDOWNLOAD_OPTIONS": set_xdownload_options,
@@ -73,6 +75,15 @@ def set_xframe(value: str):
     return {"X-Frame-Options": value}
 
 
+# Set Permissions-Policy response header
+def set_permissions_policy(value: str):
+    pattern = r"^(?:(accelerometer|autoplay|camera|clipboard-read|clipboard-write|fullscreen|geolocation|gyroscope|magnetometer|microphone|midi|payment|picture-in-picture|sync-xhr|usb|xr-spatial-tracking)=\((self)?\),?)*$"
+    match = re.match(pattern, value, re.IGNORECASE)
+    if not match:
+        value = "none"
+    return {"Permissions-Policy": value}
+
+
 # Set Referrer-Policy response header
 def set_referrer(value: str):
     pattern = r"^(no-referrer|no-referrer-when-downgrade|origin|origin-when-cross-origin|same-origin|strict-origin|strict-origin-when-cross-origin|unsafe-url)$"

backend/open_webui/utils/task.py

@@ -163,7 +163,7 @@ def emoji_generation_template(
     return template
 
 
-def search_query_generation_template(
+def query_generation_template(
     template: str, messages: list[dict], user: Optional[dict] = None
 ) -> str:
     prompt = get_last_user_message(messages)

backend/open_webui/utils/tools.py

@@ -4,7 +4,7 @@ from typing import Awaitable, Callable, get_type_hints
 
 from open_webui.apps.webui.models.tools import Tools
 from open_webui.apps.webui.models.users import UserModel
-from open_webui.apps.webui.utils import load_toolkit_module_by_id
+from open_webui.apps.webui.utils import load_tools_module_by_id
 from open_webui.utils.schemas import json_schema_to_model
 
 log = logging.getLogger(__name__)
@@ -32,15 +32,16 @@ def apply_extra_params_to_tool_function(
 def get_tools(
     webui_app, tool_ids: list[str], user: UserModel, extra_params: dict
 ) -> dict[str, dict]:
-    tools = {}
+    tools_dict = {}
+
     for tool_id in tool_ids:
-        toolkit = Tools.get_tool_by_id(tool_id)
-        if toolkit is None:
+        tools = Tools.get_tool_by_id(tool_id)
+        if tools is None:
             continue
 
         module = webui_app.state.TOOLS.get(tool_id, None)
         if module is None:
-            module, _ = load_toolkit_module_by_id(tool_id)
+            module, _ = load_tools_module_by_id(tool_id)
             webui_app.state.TOOLS[tool_id] = module
 
         extra_params["__id__"] = tool_id
@@ -53,11 +54,19 @@ def get_tools(
                 **Tools.get_user_valves_by_id_and_user_id(tool_id, user.id)
             )
 
-        for spec in toolkit.specs:
+        for spec in tools.specs:
             # TODO: Fix hack for OpenAI API
             for val in spec.get("parameters", {}).get("properties", {}).values():
                 if val["type"] == "str":
                     val["type"] = "string"
+
+            # Remove internal parameters
+            spec["parameters"]["properties"] = {
+                key: val
+                for key, val in spec["parameters"]["properties"].items()
+                if not key.startswith("__")
+            }
+
             function_name = spec["name"]
 
             # convert to function that takes only model params and inserts custom params
@@ -77,13 +86,14 @@ def get_tools(
             }
 
             # TODO: if collision, prepend toolkit name
-            if function_name in tools:
-                log.warning(f"Tool {function_name} already exists in another toolkit!")
-                log.warning(f"Collision between {toolkit} and {tool_id}.")
-                log.warning(f"Discarding {toolkit}.{function_name}")
+            if function_name in tools_dict:
+                log.warning(f"Tool {function_name} already exists in another tools!")
+                log.warning(f"Collision between {tools} and {tool_id}.")
+                log.warning(f"Discarding {tools}.{function_name}")
             else:
-                tools[function_name] = tool_dict
-    return tools
+                tools_dict[function_name] = tool_dict
+
+    return tools_dict
 
 
 def doc_to_dict(docstring):

backend/open_webui/utils/utils.py

@@ -1,12 +1,15 @@
 import logging
 import uuid
+import jwt
+
 from datetime import UTC, datetime, timedelta
-from typing import Optional, Union
+from typing import Optional, Union, List, Dict
 
-import jwt
 from open_webui.apps.webui.models.users import Users
+
 from open_webui.constants import ERROR_MESSAGES
 from open_webui.env import WEBUI_SECRET_KEY
+
 from fastapi import Depends, HTTPException, Request, Response, status
 from fastapi.security import HTTPAuthorizationCredentials, HTTPBearer
 from passlib.context import CryptContext
@@ -88,10 +91,21 @@ def get_current_user(
 
     # auth by api key
     if token.startswith("sk-"):
+        if not request.state.enable_api_key:
+            raise HTTPException(
+                status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.API_KEY_NOT_ALLOWED
+            )
         return get_current_user_by_api_key(token)
 
     # auth by jwt token
-    data = decode_token(token)
+    try:
+        data = decode_token(token)
+    except Exception as e:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="Invalid token",
+        )
+
     if data is not None and "id" in data:
         user = Users.get_user_by_id(data["id"])
         if user is None:

backend/requirements.txt

@@ -1,7 +1,7 @@
 fastapi==0.111.0
 uvicorn[standard]==0.30.6
 pydantic==2.9.2
-python-multipart==0.0.9
+python-multipart==0.0.17
 
 Flask==3.0.3
 Flask-Cors==5.0.0
@@ -13,18 +13,20 @@ passlib[bcrypt]==1.7.4
 requests==2.32.3
 aiohttp==3.10.8
 async-timeout
+aiocache
 
 sqlalchemy==2.0.32
 alembic==1.13.2
 peewee==3.17.6
 peewee-migrate==1.12.2
 psycopg2-binary==2.9.9
+pgvector==0.3.5
 PyMySQL==1.1.1
 bcrypt==4.2.0
 
 pymongo
 redis
-boto3==1.35.0
+boto3==1.35.53
 
 argon2-cffi==23.1.0
 APScheduler==3.10.4
@@ -35,14 +37,15 @@ anthropic
 google-generativeai==0.7.2
 tiktoken
 
-langchain==0.2.15
-langchain-community==0.2.12
+langchain==0.3.7
+langchain-community==0.3.7
 langchain-chroma==0.1.4
 
 fake-useragent==1.5.1
-chromadb==0.5.9
-pymilvus==2.4.7
+chromadb==0.5.15
+pymilvus==2.4.9
 qdrant-client~=1.12.0
+opensearch-py==2.7.1
 
 sentence-transformers==3.2.0
 colbert-ai==0.2.21
@@ -51,7 +54,7 @@ einops==0.8.0
 
 ftfy==6.2.3
 pypdf==4.3.1
-xhtml2pdf==0.2.16
+fpdf2==2.7.9
 pymdown-extensions==10.11.2
 docx2txt==0.8
 python-pptx==1.0.0
@@ -65,11 +68,11 @@ pyxlsb==1.0.10
 xlrd==2.0.1
 validators==0.33.0
 psutil
+sentencepiece
+soundfile==0.12.1
 
 opencv-python-headless==4.10.0.84
 rapidocr-onnxruntime==1.3.24
-
-fpdf2==2.7.9
 rank-bm25==0.2.2
 
 faster-whisper==1.0.3
@@ -79,12 +82,12 @@ authlib==1.3.2
 
 black==24.8.0
 langfuse==2.44.0
-youtube-transcript-api==0.6.2
+youtube-transcript-api==0.6.3
 pytube==15.0.0
 
 extract_msg
 pydub
-duckduckgo-search~=6.2.13
+duckduckgo-search~=6.3.5
 
 ## Tests
 docker~=7.1.0
@@ -92,3 +95,6 @@ pytest~=8.3.2
 pytest-docker~=3.1.1
 
 googleapis-common-protos==1.63.2
+
+## LDAP
+ldap3==2.9.1

package-lock.json

@@ -1,18 +1,19 @@
 {
 	"name": "open-webui",
-	"version": "0.3.35",
+	"version": "0.4.2",
 	"lockfileVersion": 3,
 	"requires": true,
 	"packages": {
 		"": {
 			"name": "open-webui",
-			"version": "0.3.35",
+			"version": "0.4.2",
 			"dependencies": {
 				"@codemirror/lang-javascript": "^6.2.2",
 				"@codemirror/lang-python": "^6.1.6",
 				"@codemirror/language-data": "^6.5.1",
 				"@codemirror/theme-one-dark": "^6.1.2",
 				"@huggingface/transformers": "^3.0.0",
+				"@mediapipe/tasks-vision": "^0.10.17",
 				"@pyscript/core": "^0.4.32",
 				"@sveltejs/adapter-node": "^2.0.0",
 				"@xyflow/svelte": "^0.1.19",
@@ -1749,6 +1750,11 @@
 				"@lezer/lr": "^1.4.0"
 			}
 		},
+		"node_modules/@mediapipe/tasks-vision": {
+			"version": "0.10.17",
+			"resolved": "https://registry.npmjs.org/@mediapipe/tasks-vision/-/tasks-vision-0.10.17.tgz",
+			"integrity": "sha512-CZWV/q6TTe8ta61cZXjfnnHsfWIdFhms03M9T7Cnd5y2mdpylJM0rF1qRq+wsQVRMLz1OYPVEBU9ph2Bx8cxrg=="
+		},
 		"node_modules/@melt-ui/svelte": {
 			"version": "0.76.0",
 			"resolved": "https://registry.npmjs.org/@melt-ui/svelte/-/svelte-0.76.0.tgz",
@@ -3993,9 +3999,10 @@
 			"integrity": "sha512-VQ2MBenTq1fWZUH9DJNGti7kKv6EeAuYr3cLwxUWhIu1baTaXh4Ib5W2CqHVqib4/MqbYGJqiL3Zb8GJZr3l4g=="
 		},
 		"node_modules/cross-spawn": {
-			"version": "7.0.3",
-			"resolved": "https://registry.npmjs.org/cross-spawn/-/cross-spawn-7.0.3.tgz",
-			"integrity": "sha512-iRDPJKUPVEND7dHPO8rkbOnPpyDygcDFtWjpeWNCgy8WP2rXcxXL8TskReQl6OrB2G7+UJrags1q15Fudc7G6w==",
+			"version": "7.0.6",
+			"resolved": "https://registry.npmjs.org/cross-spawn/-/cross-spawn-7.0.6.tgz",
+			"integrity": "sha512-uV2QOWP2nWzsy2aMp8aRibhi9dlzF5Hgh5SHaB9OiTGEyDTiJJyx0uy51QXdyWbtAHNua4XJzUKca3OzKUd3vA==",
+			"license": "MIT",
 			"dependencies": {
 				"path-key": "^3.1.0",
 				"shebang-command": "^2.0.0",

package.json

@@ -1,6 +1,6 @@
 {
 	"name": "open-webui",
-	"version": "0.3.35",
+	"version": "0.4.2",
 	"private": true,
 	"scripts": {
 		"dev": "npm run pyodide:fetch && vite dev --host",
@@ -53,6 +53,7 @@
 		"@codemirror/language-data": "^6.5.1",
 		"@codemirror/theme-one-dark": "^6.1.2",
 		"@huggingface/transformers": "^3.0.0",
+		"@mediapipe/tasks-vision": "^0.10.17",
 		"@pyscript/core": "^0.4.32",
 		"@sveltejs/adapter-node": "^2.0.0",
 		"@xyflow/svelte": "^0.1.19",

pyproject.toml

@@ -9,7 +9,7 @@ dependencies = [
     "fastapi==0.111.0",
     "uvicorn[standard]==0.30.6",
     "pydantic==2.9.2",
-    "python-multipart==0.0.9",
+    "python-multipart==0.0.17",
 
     "Flask==3.0.3",
     "Flask-Cors==5.0.0",
@@ -21,18 +21,20 @@ dependencies = [
     "requests==2.32.3",
     "aiohttp==3.10.8",
     "async-timeout",
+    "aiocache",
 
     "sqlalchemy==2.0.32",
     "alembic==1.13.2",
     "peewee==3.17.6",
     "peewee-migrate==1.12.2",
     "psycopg2-binary==2.9.9",
+    "pgvector==0.3.5",
     "PyMySQL==1.1.1",
     "bcrypt==4.2.0",
 
     "pymongo",
     "redis",
-    "boto3==1.35.0",
+    "boto3==1.35.53",
 
     "argon2-cffi==23.1.0",
     "APScheduler==3.10.4",
@@ -42,13 +44,15 @@ dependencies = [
     "google-generativeai==0.7.2",
     "tiktoken",
 
-    "langchain==0.2.15",
-    "langchain-community==0.2.12",
+    "langchain==0.3.7",
+    "langchain-community==0.3.7",
     "langchain-chroma==0.1.4",
 
     "fake-useragent==1.5.1",
-    "chromadb==0.5.9",
-    "pymilvus==2.4.7",
+    "chromadb==0.5.15",
+    "pymilvus==2.4.9",
+    "qdrant-client~=1.12.0",
+    "opensearch-py==2.7.1",
 
     "sentence-transformers==3.2.0",
     "colbert-ai==0.2.21",
@@ -56,7 +60,7 @@ dependencies = [
 
     "ftfy==6.2.3",
     "pypdf==4.3.1",
-    "xhtml2pdf==0.2.16",
+    "fpdf2==2.7.9",
     "pymdown-extensions==10.11.2",
     "docx2txt==0.8",
     "python-pptx==1.0.0",
@@ -70,11 +74,11 @@ dependencies = [
     "xlrd==2.0.1",
     "validators==0.33.0",
     "psutil",
+    "sentencepiece",
+    "soundfile==0.12.1",
 
     "opencv-python-headless==4.10.0.84",
     "rapidocr-onnxruntime==1.3.24",
-
-    "fpdf2==2.7.9",
     "rank-bm25==0.2.2",
 
     "faster-whisper==1.0.3",
@@ -84,18 +88,20 @@ dependencies = [
 
     "black==24.8.0",
     "langfuse==2.44.0",
-    "youtube-transcript-api==0.6.2",
+    "youtube-transcript-api==0.6.3",
     "pytube==15.0.0",
 
     "extract_msg",
     "pydub",
-    "duckduckgo-search~=6.2.13",
+    "duckduckgo-search~=6.3.5",
 
     "docker~=7.1.0",
     "pytest~=8.3.2",
     "pytest-docker~=3.1.1",
 
-    "googleapis-common-protos==1.63.2"
+    "googleapis-common-protos==1.63.2",
+
+    "ldap3==2.9.1"
 ]
 readme = "README.md"
 requires-python = ">= 3.11, < 3.12.0a1"

src/app.css

@@ -16,6 +16,12 @@
 	font-display: swap;
 }
 
+@font-face {
+	font-family: 'InstrumentSerif';
+	src: url('/assets/fonts/InstrumentSerif-Regular.ttf');
+	font-display: swap;
+}
+
 html {
 	word-break: break-word;
 }
@@ -26,6 +32,10 @@ code {
 	width: auto;
 }
 
+.font-secondary {
+	font-family: 'InstrumentSerif', sans-serif;
+}
+
 math {
 	margin-top: 1rem;
 }

src/lib/apis/auths/index.ts

@@ -110,6 +110,150 @@ export const getSessionUser = async (token: string) => {
 	return res;
 };
 
+export const ldapUserSignIn = async (user: string, password: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/ldap`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json'
+		},
+		credentials: 'include',
+		body: JSON.stringify({
+			user: user,
+			password: password
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getLdapConfig = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/admin/config/ldap`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const updateLdapConfig = async (token: string = '', enable_ldap: boolean) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/admin/config/ldap`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		},
+		body: JSON.stringify({
+			enable_ldap: enable_ldap
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getLdapServer = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/admin/config/ldap/server`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const updateLdapServer = async (token: string = '', body: object) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/admin/config/ldap/server`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		},
+		body: JSON.stringify(body)
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const userSignIn = async (email: string, password: string) => {
 	let error = null;
 

src/lib/apis/groups/index.ts

@@ -0,0 +1,162 @@
+import { WEBUI_API_BASE_URL } from '$lib/constants';
+
+export const createNewGroup = async (token: string, group: object) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/groups/create`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			...group
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getGroups = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/groups/`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getGroupById = async (token: string, id: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/groups/id/${id}`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const updateGroupById = async (token: string, id: string, group: object) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/groups/id/${id}/update`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			...group
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const deleteGroupById = async (token: string, id: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/groups/id/${id}/delete`, {
+		method: 'DELETE',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};

src/lib/apis/index.ts

@@ -1,9 +1,8 @@
 import { WEBUI_API_BASE_URL, WEBUI_BASE_URL } from '$lib/constants';
 
-export const getModels = async (token: string = '') => {
+export const getModels = async (token: string = '', base: boolean = false) => {
 	let error = null;
-
-	const res = await fetch(`${WEBUI_BASE_URL}/api/models`, {
+	const res = await fetch(`${WEBUI_BASE_URL}/api/models${base ? '/base' : ''}`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',
@@ -16,8 +15,8 @@ export const getModels = async (token: string = '') => {
 			return res.json();
 		})
 		.catch((err) => {
-			console.log(err);
 			error = err;
+			console.log(err);
 			return null;
 		});
 
@@ -26,26 +25,10 @@ export const getModels = async (token: string = '') => {
 	}
 
 	let models = res?.data ?? [];
-
 	models = models
 		.filter((models) => models)
 		// Sort the models
 		.sort((a, b) => {
-			// Check if models have position property
-			const aHasPosition = a.info?.meta?.position !== undefined;
-			const bHasPosition = b.info?.meta?.position !== undefined;
-
-			// If both a and b have the position property
-			if (aHasPosition && bHasPosition) {
-				return a.info.meta.position - b.info.meta.position;
-			}
-
-			// If only a has the position property, it should come first
-			if (aHasPosition) return -1;
-
-			// If only b has the position property, it should come first
-			if (bHasPosition) return 1;
-
 			// Compare case-insensitively by name for models without position property
 			const lowerA = a.name.toLowerCase();
 			const lowerB = b.name.toLowerCase();
@@ -365,15 +348,16 @@ export const generateEmoji = async (
 	return null;
 };
 
-export const generateSearchQuery = async (
+export const generateQueries = async (
 	token: string = '',
 	model: string,
 	messages: object[],
-	prompt: string
+	prompt: string,
+	type?: string = 'web_search'
 ) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_BASE_URL}/api/task/query/completions`, {
+	const res = await fetch(`${WEBUI_BASE_URL}/api/task/queries/completions`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -383,7 +367,8 @@ export const generateSearchQuery = async (
 		body: JSON.stringify({
 			model: model,
 			messages: messages,
-			prompt: prompt
+			prompt: prompt,
+			type: type
 		})
 	})
 		.then(async (res) => {
@@ -402,7 +387,39 @@ export const generateSearchQuery = async (
 		throw error;
 	}
 
-	return res?.choices[0]?.message?.content.replace(/["']/g, '') ?? prompt;
+	try {
+		// Step 1: Safely extract the response string
+		const response = res?.choices[0]?.message?.content ?? '';
+
+		// Step 2: Attempt to fix common JSON format issues like single quotes
+		const sanitizedResponse = response.replace(/['‘’`]/g, '"'); // Convert single quotes to double quotes for valid JSON
+
+		// Step 3: Find the relevant JSON block within the response
+		const jsonStartIndex = sanitizedResponse.indexOf('{');
+		const jsonEndIndex = sanitizedResponse.lastIndexOf('}');
+
+		// Step 4: Check if we found a valid JSON block (with both `{` and `}`)
+		if (jsonStartIndex !== -1 && jsonEndIndex !== -1) {
+			const jsonResponse = sanitizedResponse.substring(jsonStartIndex, jsonEndIndex + 1);
+
+			// Step 5: Parse the JSON block
+			const parsed = JSON.parse(jsonResponse);
+
+			// Step 6: If there's a "queries" key, return the queries array; otherwise, return an empty array
+			if (parsed && parsed.queries) {
+				return Array.isArray(parsed.queries) ? parsed.queries : [];
+			} else {
+				return [];
+			}
+		}
+
+		// If no valid JSON block found, return an empty array
+		return [];
+	} catch (e) {
+		// Catch and safely return empty array on any parsing errors
+		console.error('Failed to parse response: ', e);
+		return [];
+	}
 };
 
 export const generateMoACompletion = async (

src/lib/apis/knowledge/index.ts

@@ -1,6 +1,11 @@
 import { WEBUI_API_BASE_URL } from '$lib/constants';
 
-export const createNewKnowledge = async (token: string, name: string, description: string) => {
+export const createNewKnowledge = async (
+	token: string,
+	name: string,
+	description: string,
+	accessControl: null | object
+) => {
 	let error = null;
 
 	const res = await fetch(`${WEBUI_API_BASE_URL}/knowledge/create`, {
@@ -12,7 +17,8 @@ export const createNewKnowledge = async (token: string, name: string, descriptio
 		},
 		body: JSON.stringify({
 			name: name,
-			description: description
+			description: description,
+			access_control: accessControl
 		})
 	})
 		.then(async (res) => {
@@ -32,7 +38,7 @@ export const createNewKnowledge = async (token: string, name: string, descriptio
 	return res;
 };
 
-export const getKnowledgeItems = async (token: string = '') => {
+export const getKnowledgeBases = async (token: string = '') => {
 	let error = null;
 
 	const res = await fetch(`${WEBUI_API_BASE_URL}/knowledge/`, {
@@ -63,6 +69,37 @@ export const getKnowledgeItems = async (token: string = '') => {
 	return res;
 };
 
+export const getKnowledgeBaseList = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/knowledge/list`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const getKnowledgeById = async (token: string, id: string) => {
 	let error = null;
 
@@ -99,6 +136,7 @@ type KnowledgeUpdateForm = {
 	name?: string;
 	description?: string;
 	data?: object;
+	access_control?: null | object;
 };
 
 export const updateKnowledgeById = async (token: string, id: string, form: KnowledgeUpdateForm) => {
@@ -114,7 +152,8 @@ export const updateKnowledgeById = async (token: string, id: string, form: Knowl
 		body: JSON.stringify({
 			name: form?.name ? form.name : undefined,
 			description: form?.description ? form.description : undefined,
-			data: form?.data ? form.data : undefined
+			data: form?.data ? form.data : undefined,
+			access_control: form.access_control
 		})
 	})
 		.then(async (res) => {

src/lib/apis/models/index.ts

@@ -1,9 +1,71 @@
 import { WEBUI_API_BASE_URL } from '$lib/constants';
 
-export const addNewModel = async (token: string, model: object) => {
+export const getModels = async (token: string = '') => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/models/add`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getBaseModels = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/base`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const createNewModel = async (token: string, model: object) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/create`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -29,10 +91,13 @@ export const addNewModel = async (token: string, model: object) => {
 	return res;
 };
 
-export const getModelInfos = async (token: string = '') => {
+export const getModelById = async (token: string, id: string) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/models`, {
+	const searchParams = new URLSearchParams();
+	searchParams.append('id', id);
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/model?${searchParams.toString()}`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',
@@ -49,6 +114,7 @@ export const getModelInfos = async (token: string = '') => {
 		})
 		.catch((err) => {
 			error = err;
+
 			console.log(err);
 			return null;
 		});
@@ -60,14 +126,14 @@ export const getModelInfos = async (token: string = '') => {
 	return res;
 };
 
-export const getModelById = async (token: string, id: string) => {
+export const toggleModelById = async (token: string, id: string) => {
 	let error = null;
 
 	const searchParams = new URLSearchParams();
 	searchParams.append('id', id);
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/models?${searchParams.toString()}`, {
-		method: 'GET',
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/model/toggle?${searchParams.toString()}`, {
+		method: 'POST',
 		headers: {
 			Accept: 'application/json',
 			'Content-Type': 'application/json',
@@ -101,7 +167,7 @@ export const updateModelById = async (token: string, id: string, model: object)
 	const searchParams = new URLSearchParams();
 	searchParams.append('id', id);
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/models/update?${searchParams.toString()}`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/model/update?${searchParams.toString()}`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -137,7 +203,39 @@ export const deleteModelById = async (token: string, id: string) => {
 	const searchParams = new URLSearchParams();
 	searchParams.append('id', id);
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/models/delete?${searchParams.toString()}`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/model/delete?${searchParams.toString()}`, {
+		method: 'DELETE',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const deleteAllModels = async (token: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/models/delete/all`, {
 		method: 'DELETE',
 		headers: {
 			Accept: 'application/json',

src/lib/apis/ollama/index.ts

@@ -1,5 +1,40 @@
 import { OLLAMA_API_BASE_URL } from '$lib/constants';
 
+export const verifyOllamaConnection = async (
+	token: string = '',
+	url: string = '',
+	key: string = ''
+) => {
+	let error = null;
+
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/verify`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			Authorization: `Bearer ${token}`,
+			'Content-Type': 'application/json'
+		},
+		body: JSON.stringify({
+			url,
+			key
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = `Ollama: ${err?.error?.message ?? 'Network Problem'}`;
+			return [];
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const getOllamaConfig = async (token: string = '') => {
 	let error = null;
 
@@ -32,7 +67,13 @@ export const getOllamaConfig = async (token: string = '') => {
 	return res;
 };
 
-export const updateOllamaConfig = async (token: string = '', enable_ollama_api: boolean) => {
+type OllamaConfig = {
+	ENABLE_OLLAMA_API: boolean;
+	OLLAMA_BASE_URLS: string[];
+	OLLAMA_API_CONFIGS: object;
+};
+
+export const updateOllamaConfig = async (token: string = '', config: OllamaConfig) => {
 	let error = null;
 
 	const res = await fetch(`${OLLAMA_API_BASE_URL}/config/update`, {
@@ -43,7 +84,7 @@ export const updateOllamaConfig = async (token: string = '', enable_ollama_api:
 			...(token && { authorization: `Bearer ${token}` })
 		},
 		body: JSON.stringify({
-			enable_ollama_api: enable_ollama_api
+			...config
 		})
 	})
 		.then(async (res) => {
@@ -166,10 +207,10 @@ export const getOllamaVersion = async (token: string, urlIdx?: number) => {
 	return res?.version ?? false;
 };
 
-export const getOllamaModels = async (token: string = '') => {
+export const getOllamaModels = async (token: string = '', urlIdx: null | number = null) => {
 	let error = null;
 
-	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/tags`, {
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/tags${urlIdx !== null ? `/${urlIdx}` : ''}`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',

src/lib/apis/openai/index.ts

@@ -32,7 +32,14 @@ export const getOpenAIConfig = async (token: string = '') => {
 	return res;
 };
 
-export const updateOpenAIConfig = async (token: string = '', enable_openai_api: boolean) => {
+type OpenAIConfig = {
+	ENABLE_OPENAI_API: boolean;
+	OPENAI_API_BASE_URLS: string[];
+	OPENAI_API_KEYS: string[];
+	OPENAI_API_CONFIGS: object;
+};
+
+export const updateOpenAIConfig = async (token: string = '', config: OpenAIConfig) => {
 	let error = null;
 
 	const res = await fetch(`${OPENAI_API_BASE_URL}/config/update`, {
@@ -43,7 +50,7 @@ export const updateOpenAIConfig = async (token: string = '', enable_openai_api:
 			...(token && { authorization: `Bearer ${token}` })
 		},
 		body: JSON.stringify({
-			enable_openai_api: enable_openai_api
+			...config
 		})
 	})
 		.then(async (res) => {
@@ -231,41 +238,39 @@ export const getOpenAIModels = async (token: string, urlIdx?: number) => {
 	return res;
 };
 
-export const getOpenAIModelsDirect = async (
-	base_url: string = 'https://api.openai.com/v1',
-	api_key: string = ''
+export const verifyOpenAIConnection = async (
+	token: string = '',
+	url: string = 'https://api.openai.com/v1',
+	key: string = ''
 ) => {
 	let error = null;
 
-	const res = await fetch(`${base_url}/models`, {
-		method: 'GET',
+	const res = await fetch(`${OPENAI_API_BASE_URL}/verify`, {
+		method: 'POST',
 		headers: {
-			'Content-Type': 'application/json',
-			Authorization: `Bearer ${api_key}`
-		}
+			Accept: 'application/json',
+			Authorization: `Bearer ${token}`,
+			'Content-Type': 'application/json'
+		},
+		body: JSON.stringify({
+			url,
+			key
+		})
 	})
 		.then(async (res) => {
 			if (!res.ok) throw await res.json();
 			return res.json();
 		})
 		.catch((err) => {
-			console.log(err);
 			error = `OpenAI: ${err?.error?.message ?? 'Network Problem'}`;
-			return null;
+			return [];
 		});
 
 	if (error) {
 		throw error;
 	}
 
-	const models = Array.isArray(res) ? res : (res?.data ?? null);
-
-	return models
-		.map((model) => ({ id: model.id, name: model.name ?? model.id, external: true }))
-		.filter((model) => (base_url.includes('openai') ? model.name.includes('gpt') : true))
-		.sort((a, b) => {
-			return a.name.localeCompare(b.name);
-		});
+	return res;
 };
 
 export const generateOpenAIChatCompletion = async (

src/lib/apis/prompts/index.ts

@@ -1,11 +1,13 @@
 import { WEBUI_API_BASE_URL } from '$lib/constants';
 
-export const createNewPrompt = async (
-	token: string,
-	command: string,
-	title: string,
-	content: string
-) => {
+type PromptItem = {
+	command: string;
+	title: string;
+	content: string;
+	access_control: null | object;
+};
+
+export const createNewPrompt = async (token: string, prompt: PromptItem) => {
 	let error = null;
 
 	const res = await fetch(`${WEBUI_API_BASE_URL}/prompts/create`, {
@@ -16,9 +18,8 @@ export const createNewPrompt = async (
 			authorization: `Bearer ${token}`
 		},
 		body: JSON.stringify({
-			command: `/${command}`,
-			title: title,
-			content: content
+			...prompt,
+			command: `/${prompt.command}`
 		})
 	})
 		.then(async (res) => {
@@ -69,6 +70,37 @@ export const getPrompts = async (token: string = '') => {
 	return res;
 };
 
+export const getPromptList = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/prompts/list`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const getPromptByCommand = async (token: string, command: string) => {
 	let error = null;
 
@@ -101,15 +133,10 @@ export const getPromptByCommand = async (token: string, command: string) => {
 	return res;
 };
 
-export const updatePromptByCommand = async (
-	token: string,
-	command: string,
-	title: string,
-	content: string
-) => {
+export const updatePromptByCommand = async (token: string, prompt: PromptItem) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/prompts/command/${command}/update`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/prompts/command/${prompt.command}/update`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -117,9 +144,8 @@ export const updatePromptByCommand = async (
 			authorization: `Bearer ${token}`
 		},
 		body: JSON.stringify({
-			command: `/${command}`,
-			title: title,
-			content: content
+			...prompt,
+			command: `/${prompt.command}`
 		})
 	})
 		.then(async (res) => {

src/lib/apis/tools/index.ts

@@ -62,6 +62,37 @@ export const getTools = async (token: string = '') => {
 	return res;
 };
 
+export const getToolList = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/tools/list`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const exportTools = async (token: string = '') => {
 	let error = null;
 

src/lib/apis/users/index.ts

@@ -1,10 +1,10 @@
 import { WEBUI_API_BASE_URL } from '$lib/constants';
 import { getUserPosition } from '$lib/utils';
 
-export const getUserPermissions = async (token: string) => {
+export const getUserGroups = async (token: string) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/users/permissions/user`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/users/groups`, {
 		method: 'GET',
 		headers: {
 			'Content-Type': 'application/json',
@@ -28,10 +28,37 @@ export const getUserPermissions = async (token: string) => {
 	return res;
 };
 
-export const updateUserPermissions = async (token: string, permissions: object) => {
+export const getUserDefaultPermissions = async (token: string) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/users/permissions/user`, {
+	const res = await fetch(`${WEBUI_API_BASE_URL}/users/default/permissions`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const updateUserDefaultPermissions = async (token: string, permissions: object) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/users/default/permissions`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'application/json',

src/lib/components/ChangelogModal.svelte

@@ -22,7 +22,7 @@
 	});
 </script>
 
-<Modal bind:show>
+<Modal bind:show size="lg">
 	<div class="px-5 pt-4 dark:text-gray-300 text-gray-700">
 		<div class="flex justify-between items-start">
 			<div class="text-xl font-semibold">
@@ -59,7 +59,7 @@
 	</div>
 
 	<div class=" w-full p-4 px-5 text-gray-700 dark:text-gray-100">
-		<div class=" overflow-y-scroll max-h-80 scrollbar-hidden">
+		<div class=" overflow-y-scroll max-h-96 scrollbar-hidden">
 			<div class="mb-3">
 				{#if changelog}
 					{#each Object.keys(changelog) as version}
@@ -111,7 +111,7 @@
 					await updateUserSettings(localStorage.token, { ui: $settings });
 					show = false;
 				}}
-				class=" px-4 py-2 bg-emerald-700 hover:bg-emerald-800 text-gray-100 transition rounded-lg"
+				class="px-3.5 py-1.5 text-sm font-medium bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full"
 			>
 				<span class="relative">{$i18n.t("Okay, Let's Go!")}</span>
 			</button>

src/lib/components/OnBoarding.svelte

@@ -0,0 +1,78 @@
+<script>
+	import { getContext } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import { WEBUI_BASE_URL } from '$lib/constants';
+
+	import Marquee from './common/Marquee.svelte';
+	import SlideShow from './common/SlideShow.svelte';
+	import ArrowRightCircle from './icons/ArrowRightCircle.svelte';
+
+	export let show = true;
+	export let getStartedHandler = () => {};
+</script>
+
+{#if show}
+	<div class="w-full h-screen max-h-[100dvh] text-white relative">
+		<div class="fixed m-10 z-50">
+			<div class="flex space-x-2">
+				<div class=" self-center">
+					<img
+						crossorigin="anonymous"
+						src="{WEBUI_BASE_URL}/static/favicon.png"
+						class=" w-6 rounded-full"
+						alt="logo"
+					/>
+				</div>
+			</div>
+		</div>
+
+		<SlideShow duration={5000} />
+
+		<div
+			class="w-full h-full absolute top-0 left-0 bg-gradient-to-t from-20% from-black to-transparent"
+		></div>
+
+		<div class="w-full h-full absolute top-0 left-0 backdrop-blur-sm bg-black/50"></div>
+
+		<div class="relative bg-transparent w-full min-h-screen flex z-10">
+			<div class="flex flex-col justify-end w-full items-center pb-10 text-center">
+				<div class="text-5xl lg:text-7xl font-secondary">
+					<Marquee
+						duration={5000}
+						words={[
+							$i18n.t('Explore the cosmos'),
+							$i18n.t('Unlock mysteries'),
+							$i18n.t('Chart new frontiers'),
+							$i18n.t('Dive into knowledge'),
+							$i18n.t('Discover wonders'),
+							$i18n.t('Ignite curiosity'),
+							$i18n.t('Forge new paths'),
+							$i18n.t('Unravel secrets'),
+							$i18n.t('Pioneer insights'),
+							$i18n.t('Embark on adventures')
+						]}
+					/>
+
+					<div class="mt-0.5">{$i18n.t(`wherever you are`)}</div>
+				</div>
+
+				<div class="flex justify-center mt-8">
+					<div class="flex flex-col justify-center items-center">
+						<button
+							class="relative z-20 flex p-1 rounded-full bg-white/5 hover:bg-white/10 transition font-medium text-sm"
+							on:click={() => {
+								getStartedHandler();
+							}}
+						>
+							<ArrowRightCircle className="size-6" />
+						</button>
+						<div class="mt-1.5 font-primary text-base font-medium">{$i18n.t(`Get started`)}</div>
+					</div>
+				</div>
+			</div>
+
+			<!-- <div class="absolute bottom-12 left-0 right-0 w-full"></div> -->
+		</div>
+	</div>
+{/if}

src/lib/components/admin/Evaluations.svelte

@@ -1,677 +1,100 @@
-<script lang="ts">
-	import fileSaver from 'file-saver';
-	const { saveAs } = fileSaver;
-
-	import { onMount, getContext } from 'svelte';
-	import dayjs from 'dayjs';
-	import relativeTime from 'dayjs/plugin/relativeTime';
-	dayjs.extend(relativeTime);
-
-	import * as ort from 'onnxruntime-web';
-	import { AutoModel, AutoTokenizer } from '@huggingface/transformers';
-
-	const EMBEDDING_MODEL = 'TaylorAI/bge-micro-v2';
-	let tokenizer = null;
-	let model = null;
-
-	import { models } from '$lib/stores';
-	import { deleteFeedbackById, exportAllFeedbacks, getAllFeedbacks } from '$lib/apis/evaluations';
-
-	import FeedbackMenu from './Evaluations/FeedbackMenu.svelte';
-	import EllipsisHorizontal from '../icons/EllipsisHorizontal.svelte';
-	import Tooltip from '../common/Tooltip.svelte';
-	import Badge from '../common/Badge.svelte';
-	import Pagination from '../common/Pagination.svelte';
-	import MagnifyingGlass from '../icons/MagnifyingGlass.svelte';
-	import Share from '../icons/Share.svelte';
-	import CloudArrowUp from '../icons/CloudArrowUp.svelte';
+<script>
+	import { getContext, tick, onMount } from 'svelte';
 	import { toast } from 'svelte-sonner';
-	import Spinner from '../common/Spinner.svelte';
-	import DocumentArrowUpSolid from '../icons/DocumentArrowUpSolid.svelte';
-	import DocumentArrowDown from '../icons/DocumentArrowDown.svelte';
-	import ArrowDownTray from '../icons/ArrowDownTray.svelte';
-
-	const i18n = getContext('i18n');
+	import Leaderboard from './Evaluations/Leaderboard.svelte';
+	import Feedbacks from './Evaluations/Feedbacks.svelte';
 
-	let rankedModels = [];
-	let feedbacks = [];
+	import { getAllFeedbacks } from '$lib/apis/evaluations';
 
-	let query = '';
-	let page = 1;
+	const i18n = getContext('i18n');
 
-	let tagEmbeddings = new Map();
+	let selectedTab = 'leaderboard';
 
 	let loaded = false;
-	let loadingLeaderboard = true;
-	let debounceTimer;
-
-	$: paginatedFeedbacks = feedbacks.slice((page - 1) * 10, page * 10);
-
-	type Feedback = {
-		id: string;
-		data: {
-			rating: number;
-			model_id: string;
-			sibling_model_ids: string[] | null;
-			reason: string;
-			comment: string;
-			tags: string[];
-		};
-		user: {
-			name: string;
-			profile_image_url: string;
-		};
-		updated_at: number;
-	};
-
-	type ModelStats = {
-		rating: number;
-		won: number;
-		lost: number;
-	};
-
-	//////////////////////
-	//
-	// Rank models by Elo rating
-	//
-	//////////////////////
-
-	const rankHandler = async (similarities: Map<string, number> = new Map()) => {
-		const modelStats = calculateModelStats(feedbacks, similarities);
-
-		rankedModels = $models
-			.filter((m) => m?.owned_by !== 'arena' && (m?.info?.meta?.hidden ?? false) !== true)
-			.map((model) => {
-				const stats = modelStats.get(model.id);
-				return {
-					...model,
-					rating: stats ? Math.round(stats.rating) : '-',
-					stats: {
-						count: stats ? stats.won + stats.lost : 0,
-						won: stats ? stats.won.toString() : '-',
-						lost: stats ? stats.lost.toString() : '-'
-					}
-				};
-			})
-			.sort((a, b) => {
-				if (a.rating === '-' && b.rating !== '-') return 1;
-				if (b.rating === '-' && a.rating !== '-') return -1;
-				if (a.rating !== '-' && b.rating !== '-') return b.rating - a.rating;
-				return a.name.localeCompare(b.name);
-			});
-
-		loadingLeaderboard = false;
-	};
-
-	function calculateModelStats(
-		feedbacks: Feedback[],
-		similarities: Map<string, number>
-	): Map<string, ModelStats> {
-		const stats = new Map<string, ModelStats>();
-		const K = 32;
-
-		function getOrDefaultStats(modelId: string): ModelStats {
-			return stats.get(modelId) || { rating: 1000, won: 0, lost: 0 };
-		}
-
-		function updateStats(modelId: string, ratingChange: number, outcome: number) {
-			const currentStats = getOrDefaultStats(modelId);
-			currentStats.rating += ratingChange;
-			if (outcome === 1) currentStats.won++;
-			else if (outcome === 0) currentStats.lost++;
-			stats.set(modelId, currentStats);
-		}
-
-		function calculateEloChange(
-			ratingA: number,
-			ratingB: number,
-			outcome: number,
-			similarity: number
-		): number {
-			const expectedScore = 1 / (1 + Math.pow(10, (ratingB - ratingA) / 400));
-			return K * (outcome - expectedScore) * similarity;
-		}
-
-		feedbacks.forEach((feedback) => {
-			const modelA = feedback.data.model_id;
-			const statsA = getOrDefaultStats(modelA);
-			let outcome: number;
-
-			switch (feedback.data.rating.toString()) {
-				case '1':
-					outcome = 1;
-					break;
-				case '-1':
-					outcome = 0;
-					break;
-				default:
-					return; // Skip invalid ratings
-			}
-
-			// If the query is empty, set similarity to 1, else get the similarity from the map
-			const similarity = query !== '' ? similarities.get(feedback.id) || 0 : 1;
-			const opponents = feedback.data.sibling_model_ids || [];
-
-			opponents.forEach((modelB) => {
-				const statsB = getOrDefaultStats(modelB);
-				const changeA = calculateEloChange(statsA.rating, statsB.rating, outcome, similarity);
-				const changeB = calculateEloChange(statsB.rating, statsA.rating, 1 - outcome, similarity);
-
-				updateStats(modelA, changeA, outcome);
-				updateStats(modelB, changeB, 1 - outcome);
-			});
-		});
-
-		return stats;
-	}
-
-	//////////////////////
-	//
-	// Calculate cosine similarity
-	//
-	//////////////////////
-
-	const cosineSimilarity = (vecA, vecB) => {
-		// Ensure the lengths of the vectors are the same
-		if (vecA.length !== vecB.length) {
-			throw new Error('Vectors must be the same length');
-		}
-
-		// Calculate the dot product
-		let dotProduct = 0;
-		let normA = 0;
-		let normB = 0;
-
-		for (let i = 0; i < vecA.length; i++) {
-			dotProduct += vecA[i] * vecB[i];
-			normA += vecA[i] ** 2;
-			normB += vecB[i] ** 2;
-		}
-
-		// Calculate the magnitudes
-		normA = Math.sqrt(normA);
-		normB = Math.sqrt(normB);
-
-		// Avoid division by zero
-		if (normA === 0 || normB === 0) {
-			return 0;
-		}
-
-		// Return the cosine similarity
-		return dotProduct / (normA * normB);
-	};
-
-	const calculateMaxSimilarity = (queryEmbedding, tagEmbeddings: Map<string, number[]>) => {
-		let maxSimilarity = 0;
-		for (const tagEmbedding of tagEmbeddings.values()) {
-			const similarity = cosineSimilarity(queryEmbedding, tagEmbedding);
-			maxSimilarity = Math.max(maxSimilarity, similarity);
-		}
-		return maxSimilarity;
-	};
-
-	//////////////////////
-	//
-	// Embedding functions
-	//
-	//////////////////////
-
-	const getEmbeddings = async (text: string) => {
-		const tokens = await tokenizer(text);
-		const output = await model(tokens);
-
-		// Perform mean pooling on the last hidden states
-		const embeddings = output.last_hidden_state.mean(1);
-		return embeddings.ort_tensor.data;
-	};
-
-	const getTagEmbeddings = async (tags: string[]) => {
-		const embeddings = new Map();
-		for (const tag of tags) {
-			if (!tagEmbeddings.has(tag)) {
-				tagEmbeddings.set(tag, await getEmbeddings(tag));
-			}
-			embeddings.set(tag, tagEmbeddings.get(tag));
-		}
-		return embeddings;
-	};
-
-	const debouncedQueryHandler = async () => {
-		loadingLeaderboard = true;
-
-		if (query.trim() === '') {
-			rankHandler();
-			return;
-		}
-
-		clearTimeout(debounceTimer);
-
-		debounceTimer = setTimeout(async () => {
-			const queryEmbedding = await getEmbeddings(query);
-			const similarities = new Map<string, number>();
-
-			for (const feedback of feedbacks) {
-				const feedbackTags = feedback.data.tags || [];
-				const tagEmbeddings = await getTagEmbeddings(feedbackTags);
-				const maxSimilarity = calculateMaxSimilarity(queryEmbedding, tagEmbeddings);
-				similarities.set(feedback.id, maxSimilarity);
-			}
-
-			rankHandler(similarities);
-		}, 1500); // Debounce for 1.5 seconds
-	};
-
-	$: query, debouncedQueryHandler();
-
-	//////////////////////
-	//
-	// CRUD operations
-	//
-	//////////////////////
-
-	const deleteFeedbackHandler = async (feedbackId: string) => {
-		const response = await deleteFeedbackById(localStorage.token, feedbackId).catch((err) => {
-			toast.error(err);
-			return null;
-		});
-		if (response) {
-			feedbacks = feedbacks.filter((f) => f.id !== feedbackId);
-		}
-	};
-
-	const shareHandler = async () => {
-		toast.success($i18n.t('Redirecting you to OpenWebUI Community'));
-
-		// remove snapshot from feedbacks
-		const feedbacksToShare = feedbacks.map((f) => {
-			const { snapshot, user, ...rest } = f;
-			return rest;
-		});
-		console.log(feedbacksToShare);
-
-		const url = 'https://openwebui.com';
-		const tab = await window.open(`${url}/leaderboard`, '_blank');
-
-		// Define the event handler function
-		const messageHandler = (event) => {
-			if (event.origin !== url) return;
-			if (event.data === 'loaded') {
-				tab.postMessage(JSON.stringify(feedbacksToShare), '*');
-
-				// Remove the event listener after handling the message
-				window.removeEventListener('message', messageHandler);
-			}
-		};
-
-		window.addEventListener('message', messageHandler, false);
-	};
-
-	const exportHandler = async () => {
-		const _feedbacks = await exportAllFeedbacks(localStorage.token).catch((err) => {
-			toast.error(err);
-			return null;
-		});
-
-		if (_feedbacks) {
-			let blob = new Blob([JSON.stringify(_feedbacks)], {
-				type: 'application/json'
-			});
-			saveAs(blob, `feedback-history-export-${Date.now()}.json`);
-		}
-	};
-
-	const loadEmbeddingModel = async () => {
-		// Check if the tokenizer and model are already loaded and stored in the window object
-		if (!window.tokenizer) {
-			window.tokenizer = await AutoTokenizer.from_pretrained(EMBEDDING_MODEL);
-		}
-
-		if (!window.model) {
-			window.model = await AutoModel.from_pretrained(EMBEDDING_MODEL);
-		}
-
-		// Use the tokenizer and model from the window object
-		tokenizer = window.tokenizer;
-		model = window.model;
-
-		// Pre-compute embeddings for all unique tags
-		const allTags = new Set(feedbacks.flatMap((feedback) => feedback.data.tags || []));
-		await getTagEmbeddings(Array.from(allTags));
-	};
+	let feedbacks = [];
 
 	onMount(async () => {
 		feedbacks = await getAllFeedbacks(localStorage.token);
 		loaded = true;
 
-		rankHandler();
+		const containerElement = document.getElementById('users-tabs-container');
+
+		if (containerElement) {
+			containerElement.addEventListener('wheel', function (event) {
+				if (event.deltaY !== 0) {
+					// Adjust horizontal scroll position based on vertical scroll
+					containerElement.scrollLeft += event.deltaY;
+				}
+			});
+		}
 	});
 </script>
 
 {#if loaded}
-	<div class="mt-0.5 mb-2 gap-1 flex flex-col md:flex-row justify-between">
-		<div class="flex md:self-center text-lg font-medium px-0.5 shrink-0 items-center">
-			<div class=" gap-1">
-				{$i18n.t('Leaderboard')}
-			</div>
-
-			<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
-
-			<span class="text-lg font-medium text-gray-500 dark:text-gray-300 mr-1.5"
-				>{rankedModels.length}</span
+	<div class="flex flex-col lg:flex-row w-full h-full pb-2 lg:space-x-4">
+		<div
+			id="users-tabs-container"
+			class="tabs flex flex-row overflow-x-auto gap-2.5 max-w-full lg:gap-1 lg:flex-col lg:flex-none lg:w-40 dark:text-gray-200 text-sm font-medium text-left scrollbar-none"
+		>
+			<button
+				class="px-0.5 py-1 min-w-fit rounded-lg lg:flex-none flex text-right transition {selectedTab ===
+				'leaderboard'
+					? ''
+					: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+				on:click={() => {
+					selectedTab = 'leaderboard';
+				}}
 			>
-		</div>
-
-		<div class=" flex space-x-2">
-			<Tooltip content={$i18n.t('Re-rank models by topic similarity')}>
-				<div class="flex flex-1">
-					<div class=" self-center ml-1 mr-3">
-						<MagnifyingGlass className="size-3" />
-					</div>
-					<input
-						class=" w-full text-sm pr-4 py-1 rounded-r-xl outline-none bg-transparent"
-						bind:value={query}
-						placeholder={$i18n.t('Search')}
-						on:focus={() => {
-							loadEmbeddingModel();
-						}}
-					/>
-				</div>
-			</Tooltip>
-		</div>
-	</div>
-
-	<div
-		class="scrollbar-hidden relative whitespace-nowrap overflow-x-auto max-w-full rounded pt-0.5"
-	>
-		{#if loadingLeaderboard}
-			<div class=" absolute top-0 bottom-0 left-0 right-0 flex">
-				<div class="m-auto">
-					<Spinner />
+				<div class=" self-center mr-2">
+					<svg
+						xmlns="http://www.w3.org/2000/svg"
+						viewBox="0 0 16 16"
+						fill="currentColor"
+						class="size-4"
+					>
+						<path
+							fill-rule="evenodd"
+							d="M4 2a1.5 1.5 0 0 0-1.5 1.5v9A1.5 1.5 0 0 0 4 14h8a1.5 1.5 0 0 0 1.5-1.5V6.621a1.5 1.5 0 0 0-.44-1.06L9.94 2.439A1.5 1.5 0 0 0 8.878 2H4Zm6 5.75a.75.75 0 0 1 1.5 0v3.5a.75.75 0 0 1-1.5 0v-3.5Zm-2.75 1.5a.75.75 0 0 1 1.5 0v2a.75.75 0 0 1-1.5 0v-2Zm-2 .75a.75.75 0 0 0-.75.75v.5a.75.75 0 0 0 1.5 0v-.5a.75.75 0 0 0-.75-.75Z"
+							clip-rule="evenodd"
+						/>
+					</svg>
 				</div>
-			</div>
-		{/if}
-		{#if (rankedModels ?? []).length === 0}
-			<div class="text-center text-xs text-gray-500 dark:text-gray-400 py-1">
-				{$i18n.t('No models found')}
-			</div>
-		{:else}
-			<table
-				class="w-full text-sm text-left text-gray-500 dark:text-gray-400 table-auto max-w-full rounded {loadingLeaderboard
-					? 'opacity-20'
-					: ''}"
+				<div class=" self-center">{$i18n.t('Leaderboard')}</div>
+			</button>
+
+			<button
+				class="px-0.5 py-1 min-w-fit rounded-lg lg:flex-none flex text-right transition {selectedTab ===
+				'feedbacks'
+					? ''
+					: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+				on:click={() => {
+					selectedTab = 'feedbacks';
+				}}
 			>
-				<thead
-					class="text-xs text-gray-700 uppercase bg-gray-50 dark:bg-gray-850 dark:text-gray-400 -translate-y-0.5"
-				>
-					<tr class="">
-						<th scope="col" class="px-3 py-1.5 cursor-pointer select-none w-3">
-							{$i18n.t('RK')}
-						</th>
-						<th scope="col" class="px-3 py-1.5 cursor-pointer select-none">
-							{$i18n.t('Model')}
-						</th>
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-fit">
-							{$i18n.t('Rating')}
-						</th>
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-5">
-							{$i18n.t('Won')}
-						</th>
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-5">
-							{$i18n.t('Lost')}
-						</th>
-					</tr>
-				</thead>
-				<tbody class="">
-					{#each rankedModels as model, modelIdx (model.id)}
-						<tr class="bg-white dark:bg-gray-900 dark:border-gray-850 text-xs group">
-							<td class="px-3 py-1.5 text-left font-medium text-gray-900 dark:text-white w-fit">
-								<div class=" line-clamp-1">
-									{model?.rating !== '-' ? modelIdx + 1 : '-'}
-								</div>
-							</td>
-							<td class="px-3 py-1.5 flex flex-col justify-center">
-								<div class="flex items-center gap-2">
-									<div class="flex-shrink-0">
-										<img
-											src={model?.info?.meta?.profile_image_url ?? '/favicon.png'}
-											alt={model.name}
-											class="size-5 rounded-full object-cover shrink-0"
-										/>
-									</div>
-
-									<div class="font-medium text-gray-800 dark:text-gray-200 pr-4">
-										{model.name}
-									</div>
-								</div>
-							</td>
-							<td class="px-3 py-1.5 text-right font-medium text-gray-900 dark:text-white w-max">
-								{model.rating}
-							</td>
-
-							<td class=" px-3 py-1.5 text-right font-semibold text-green-500">
-								<div class=" w-10">
-									{#if model.stats.won === '-'}
-										-
-									{:else}
-										<span class="hidden group-hover:inline"
-											>{((model.stats.won / model.stats.count) * 100).toFixed(1)}%</span
-										>
-										<span class=" group-hover:hidden">{model.stats.won}</span>
-									{/if}
-								</div>
-							</td>
-
-							<td class="px-3 py-1.5 text-right font-semibold text-red-500">
-								<div class=" w-10">
-									{#if model.stats.lost === '-'}
-										-
-									{:else}
-										<span class="hidden group-hover:inline"
-											>{((model.stats.lost / model.stats.count) * 100).toFixed(1)}%</span
-										>
-										<span class=" group-hover:hidden">{model.stats.lost}</span>
-									{/if}
-								</div>
-							</td>
-						</tr>
-					{/each}
-				</tbody>
-			</table>
-		{/if}
-	</div>
-
-	<div class=" text-gray-500 text-xs mt-1.5 w-full flex justify-end">
-		<div class=" text-right">
-			<div class="line-clamp-1">
-				ⓘ {$i18n.t(
-					'The evaluation leaderboard is based on the Elo rating system and is updated in real-time.'
-				)}
-			</div>
-			{$i18n.t(
-				'The leaderboard is currently in beta, and we may adjust the rating calculations as we refine the algorithm.'
-			)}
-		</div>
-	</div>
-
-	<div class="pb-4"></div>
-
-	<div class="mt-0.5 mb-2 gap-1 flex flex-col md:flex-row justify-between">
-		<div class="flex md:self-center text-lg font-medium px-0.5">
-			{$i18n.t('Feedback History')}
-
-			<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
-
-			<span class="text-lg font-medium text-gray-500 dark:text-gray-300">{feedbacks.length}</span>
-		</div>
-
-		<div>
-			<div>
-				<Tooltip content={$i18n.t('Export')}>
-					<button
-						class=" p-2 rounded-xl hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 transition font-medium text-sm flex items-center space-x-1"
-						on:click={() => {
-							exportHandler();
-						}}
+				<div class=" self-center mr-2">
+					<svg
+						xmlns="http://www.w3.org/2000/svg"
+						viewBox="0 0 16 16"
+						fill="currentColor"
+						class="size-4"
 					>
-						<ArrowDownTray className="size-3" />
-					</button>
-				</Tooltip>
-			</div>
+						<path
+							fill-rule="evenodd"
+							d="M5.25 2A2.25 2.25 0 0 0 3 4.25v9a.75.75 0 0 0 1.183.613l1.692-1.195 1.692 1.195a.75.75 0 0 0 .866 0l1.692-1.195 1.693 1.195A.75.75 0 0 0 13 13.25v-9A2.25 2.25 0 0 0 10.75 2h-5.5Zm3.03 3.28a.75.75 0 0 0-1.06-1.06L4.97 6.47a.75.75 0 0 0 0 1.06l2.25 2.25a.75.75 0 0 0 1.06-1.06l-.97-.97h1.315c.76 0 1.375.616 1.375 1.375a.75.75 0 0 0 1.5 0A2.875 2.875 0 0 0 8.625 6.25H7.311l.97-.97Z"
+							clip-rule="evenodd"
+						/>
+					</svg>
+				</div>
+				<div class=" self-center">{$i18n.t('Feedbacks')}</div>
+			</button>
 		</div>
-	</div>
-
-	<div
-		class="scrollbar-hidden relative whitespace-nowrap overflow-x-auto max-w-full rounded pt-0.5"
-	>
-		{#if (feedbacks ?? []).length === 0}
-			<div class="text-center text-xs text-gray-500 dark:text-gray-400 py-1">
-				{$i18n.t('No feedbacks found')}
-			</div>
-		{:else}
-			<table
-				class="w-full text-sm text-left text-gray-500 dark:text-gray-400 table-auto max-w-full rounded"
-			>
-				<thead
-					class="text-xs text-gray-700 uppercase bg-gray-50 dark:bg-gray-850 dark:text-gray-400 -translate-y-0.5"
-				>
-					<tr class="">
-						<th scope="col" class="px-3 text-right cursor-pointer select-none w-0">
-							{$i18n.t('User')}
-						</th>
 
-						<th scope="col" class="px-3 pr-1.5 cursor-pointer select-none">
-							{$i18n.t('Models')}
-						</th>
-
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-fit">
-							{$i18n.t('Result')}
-						</th>
-
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-0">
-							{$i18n.t('Updated At')}
-						</th>
-
-						<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-0"> </th>
-					</tr>
-				</thead>
-				<tbody class="">
-					{#each paginatedFeedbacks as feedback (feedback.id)}
-						<tr class="bg-white dark:bg-gray-900 dark:border-gray-850 text-xs">
-							<td class=" py-0.5 text-right font-semibold">
-								<div class="flex justify-center">
-									<Tooltip content={feedback?.user?.name}>
-										<div class="flex-shrink-0">
-											<img
-												src={feedback?.user?.profile_image_url ?? '/user.png'}
-												alt={feedback?.user?.name}
-												class="size-5 rounded-full object-cover shrink-0"
-											/>
-										</div>
-									</Tooltip>
-								</div>
-							</td>
-
-							<td class=" py-1 pl-3 flex flex-col">
-								<div class="flex flex-col items-start gap-0.5 h-full">
-									<div class="flex flex-col h-full">
-										{#if feedback.data?.sibling_model_ids}
-											<div class="font-semibold text-gray-600 dark:text-gray-400 flex-1">
-												{feedback.data?.model_id}
-											</div>
-
-											<Tooltip content={feedback.data.sibling_model_ids.join(', ')}>
-												<div class=" text-[0.65rem] text-gray-600 dark:text-gray-400 line-clamp-1">
-													{#if feedback.data.sibling_model_ids.length > 2}
-														<!-- {$i18n.t('and {{COUNT}} more')} -->
-														{feedback.data.sibling_model_ids.slice(0, 2).join(', ')}, {$i18n.t(
-															'and {{COUNT}} more',
-															{ COUNT: feedback.data.sibling_model_ids.length - 2 }
-														)}
-													{:else}
-														{feedback.data.sibling_model_ids.join(', ')}
-													{/if}
-												</div>
-											</Tooltip>
-										{:else}
-											<div
-												class=" text-sm font-medium text-gray-600 dark:text-gray-400 flex-1 py-1.5"
-											>
-												{feedback.data?.model_id}
-											</div>
-										{/if}
-									</div>
-								</div>
-							</td>
-							<td class="px-3 py-1 text-right font-medium text-gray-900 dark:text-white w-max">
-								<div class=" flex justify-end">
-									{#if feedback.data.rating.toString() === '1'}
-										<Badge type="info" content={$i18n.t('Won')} />
-									{:else if feedback.data.rating.toString() === '0'}
-										<Badge type="muted" content={$i18n.t('Draw')} />
-									{:else if feedback.data.rating.toString() === '-1'}
-										<Badge type="error" content={$i18n.t('Lost')} />
-									{/if}
-								</div>
-							</td>
-
-							<td class=" px-3 py-1 text-right font-medium">
-								{dayjs(feedback.updated_at * 1000).fromNow()}
-							</td>
-
-							<td class=" px-3 py-1 text-right font-semibold">
-								<FeedbackMenu
-									on:delete={(e) => {
-										deleteFeedbackHandler(feedback.id);
-									}}
-								>
-									<button
-										class="self-center w-fit text-sm p-1.5 dark:text-gray-300 dark:hover:text-white hover:bg-black/5 dark:hover:bg-white/5 rounded-xl"
-									>
-										<EllipsisHorizontal />
-									</button>
-								</FeedbackMenu>
-							</td>
-						</tr>
-					{/each}
-				</tbody>
-			</table>
-		{/if}
-	</div>
-
-	{#if feedbacks.length > 0}
-		<div class=" flex flex-col justify-end w-full text-right gap-1">
-			<div class="line-clamp-1 text-gray-500 text-xs">
-				{$i18n.t('Help us create the best community leaderboard by sharing your feedback history!')}
-			</div>
-
-			<div class="flex space-x-1 ml-auto">
-				<Tooltip
-					content={$i18n.t(
-						'To protect your privacy, only ratings, model IDs, tags, and metadata are shared from your feedback—your chat logs remain private and are not included.'
-					)}
-				>
-					<button
-						class="flex text-xs items-center px-3 py-1.5 rounded-xl bg-gray-50 hover:bg-gray-100 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-200 transition"
-						on:click={async () => {
-							shareHandler();
-						}}
-					>
-						<div class=" self-center mr-2 font-medium line-clamp-1">
-							{$i18n.t('Share to OpenWebUI Community')}
-						</div>
-
-						<div class=" self-center">
-							<CloudArrowUp className="size-3" strokeWidth="3" />
-						</div>
-					</button>
-				</Tooltip>
-			</div>
+		<div class="flex-1 mt-1 lg:mt-0 overflow-y-scroll">
+			{#if selectedTab === 'leaderboard'}
+				<Leaderboard {feedbacks} />
+			{:else if selectedTab === 'feedbacks'}
+				<Feedbacks {feedbacks} />
+			{/if}
 		</div>
-	{/if}
-
-	{#if feedbacks.length > 10}
-		<Pagination bind:page count={feedbacks.length} perPage={10} />
-	{/if}
-
-	<div class="pb-12"></div>
+	</div>
 {/if}

src/lib/components/admin/Evaluations/Feedbacks.svelte

@@ -0,0 +1,283 @@
+<script lang="ts">
+	import { toast } from 'svelte-sonner';
+	import fileSaver from 'file-saver';
+	const { saveAs } = fileSaver;
+
+	import dayjs from 'dayjs';
+	import relativeTime from 'dayjs/plugin/relativeTime';
+	dayjs.extend(relativeTime);
+
+	import { onMount, getContext } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import { deleteFeedbackById, exportAllFeedbacks, getAllFeedbacks } from '$lib/apis/evaluations';
+
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import ArrowDownTray from '$lib/components/icons/ArrowDownTray.svelte';
+	import Badge from '$lib/components/common/Badge.svelte';
+	import CloudArrowUp from '$lib/components/icons/CloudArrowUp.svelte';
+	import Pagination from '$lib/components/common/Pagination.svelte';
+	import FeedbackMenu from './FeedbackMenu.svelte';
+	import EllipsisHorizontal from '$lib/components/icons/EllipsisHorizontal.svelte';
+
+	export let feedbacks = [];
+
+	let page = 1;
+	$: paginatedFeedbacks = feedbacks.slice((page - 1) * 10, page * 10);
+
+	type Feedback = {
+		id: string;
+		data: {
+			rating: number;
+			model_id: string;
+			sibling_model_ids: string[] | null;
+			reason: string;
+			comment: string;
+			tags: string[];
+		};
+		user: {
+			name: string;
+			profile_image_url: string;
+		};
+		updated_at: number;
+	};
+
+	type ModelStats = {
+		rating: number;
+		won: number;
+		lost: number;
+	};
+
+	//////////////////////
+	//
+	// CRUD operations
+	//
+	//////////////////////
+
+	const deleteFeedbackHandler = async (feedbackId: string) => {
+		const response = await deleteFeedbackById(localStorage.token, feedbackId).catch((err) => {
+			toast.error(err);
+			return null;
+		});
+		if (response) {
+			feedbacks = feedbacks.filter((f) => f.id !== feedbackId);
+		}
+	};
+
+	const shareHandler = async () => {
+		toast.success($i18n.t('Redirecting you to OpenWebUI Community'));
+
+		// remove snapshot from feedbacks
+		const feedbacksToShare = feedbacks.map((f) => {
+			const { snapshot, user, ...rest } = f;
+			return rest;
+		});
+		console.log(feedbacksToShare);
+
+		const url = 'https://openwebui.com';
+		const tab = await window.open(`${url}/leaderboard`, '_blank');
+
+		// Define the event handler function
+		const messageHandler = (event) => {
+			if (event.origin !== url) return;
+			if (event.data === 'loaded') {
+				tab.postMessage(JSON.stringify(feedbacksToShare), '*');
+
+				// Remove the event listener after handling the message
+				window.removeEventListener('message', messageHandler);
+			}
+		};
+
+		window.addEventListener('message', messageHandler, false);
+	};
+
+	const exportHandler = async () => {
+		const _feedbacks = await exportAllFeedbacks(localStorage.token).catch((err) => {
+			toast.error(err);
+			return null;
+		});
+
+		if (_feedbacks) {
+			let blob = new Blob([JSON.stringify(_feedbacks)], {
+				type: 'application/json'
+			});
+			saveAs(blob, `feedback-history-export-${Date.now()}.json`);
+		}
+	};
+</script>
+
+<div class="mt-0.5 mb-2 gap-1 flex flex-row justify-between">
+	<div class="flex md:self-center text-lg font-medium px-0.5">
+		{$i18n.t('Feedback History')}
+
+		<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
+
+		<span class="text-lg font-medium text-gray-500 dark:text-gray-300">{feedbacks.length}</span>
+	</div>
+
+	<div>
+		<div>
+			<Tooltip content={$i18n.t('Export')}>
+				<button
+					class=" p-2 rounded-xl hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 transition font-medium text-sm flex items-center space-x-1"
+					on:click={() => {
+						exportHandler();
+					}}
+				>
+					<ArrowDownTray className="size-3" />
+				</button>
+			</Tooltip>
+		</div>
+	</div>
+</div>
+
+<div class="scrollbar-hidden relative whitespace-nowrap overflow-x-auto max-w-full rounded pt-0.5">
+	{#if (feedbacks ?? []).length === 0}
+		<div class="text-center text-xs text-gray-500 dark:text-gray-400 py-1">
+			{$i18n.t('No feedbacks found')}
+		</div>
+	{:else}
+		<table
+			class="w-full text-sm text-left text-gray-500 dark:text-gray-400 table-auto max-w-full rounded"
+		>
+			<thead
+				class="text-xs text-gray-700 uppercase bg-gray-50 dark:bg-gray-850 dark:text-gray-400 -translate-y-0.5"
+			>
+				<tr class="">
+					<th scope="col" class="px-3 text-right cursor-pointer select-none w-0">
+						{$i18n.t('User')}
+					</th>
+
+					<th scope="col" class="px-3 pr-1.5 cursor-pointer select-none">
+						{$i18n.t('Models')}
+					</th>
+
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-fit">
+						{$i18n.t('Result')}
+					</th>
+
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-0">
+						{$i18n.t('Updated At')}
+					</th>
+
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-0"> </th>
+				</tr>
+			</thead>
+			<tbody class="">
+				{#each paginatedFeedbacks as feedback (feedback.id)}
+					<tr class="bg-white dark:bg-gray-900 dark:border-gray-850 text-xs">
+						<td class=" py-0.5 text-right font-semibold">
+							<div class="flex justify-center">
+								<Tooltip content={feedback?.user?.name}>
+									<div class="flex-shrink-0">
+										<img
+											src={feedback?.user?.profile_image_url ?? '/user.png'}
+											alt={feedback?.user?.name}
+											class="size-5 rounded-full object-cover shrink-0"
+										/>
+									</div>
+								</Tooltip>
+							</div>
+						</td>
+
+						<td class=" py-1 pl-3 flex flex-col">
+							<div class="flex flex-col items-start gap-0.5 h-full">
+								<div class="flex flex-col h-full">
+									{#if feedback.data?.sibling_model_ids}
+										<div class="font-semibold text-gray-600 dark:text-gray-400 flex-1">
+											{feedback.data?.model_id}
+										</div>
+
+										<Tooltip content={feedback.data.sibling_model_ids.join(', ')}>
+											<div class=" text-[0.65rem] text-gray-600 dark:text-gray-400 line-clamp-1">
+												{#if feedback.data.sibling_model_ids.length > 2}
+													<!-- {$i18n.t('and {{COUNT}} more')} -->
+													{feedback.data.sibling_model_ids.slice(0, 2).join(', ')}, {$i18n.t(
+														'and {{COUNT}} more',
+														{ COUNT: feedback.data.sibling_model_ids.length - 2 }
+													)}
+												{:else}
+													{feedback.data.sibling_model_ids.join(', ')}
+												{/if}
+											</div>
+										</Tooltip>
+									{:else}
+										<div
+											class=" text-sm font-medium text-gray-600 dark:text-gray-400 flex-1 py-1.5"
+										>
+											{feedback.data?.model_id}
+										</div>
+									{/if}
+								</div>
+							</div>
+						</td>
+						<td class="px-3 py-1 text-right font-medium text-gray-900 dark:text-white w-max">
+							<div class=" flex justify-end">
+								{#if feedback.data.rating.toString() === '1'}
+									<Badge type="info" content={$i18n.t('Won')} />
+								{:else if feedback.data.rating.toString() === '0'}
+									<Badge type="muted" content={$i18n.t('Draw')} />
+								{:else if feedback.data.rating.toString() === '-1'}
+									<Badge type="error" content={$i18n.t('Lost')} />
+								{/if}
+							</div>
+						</td>
+
+						<td class=" px-3 py-1 text-right font-medium">
+							{dayjs(feedback.updated_at * 1000).fromNow()}
+						</td>
+
+						<td class=" px-3 py-1 text-right font-semibold">
+							<FeedbackMenu
+								on:delete={(e) => {
+									deleteFeedbackHandler(feedback.id);
+								}}
+							>
+								<button
+									class="self-center w-fit text-sm p-1.5 dark:text-gray-300 dark:hover:text-white hover:bg-black/5 dark:hover:bg-white/5 rounded-xl"
+								>
+									<EllipsisHorizontal />
+								</button>
+							</FeedbackMenu>
+						</td>
+					</tr>
+				{/each}
+			</tbody>
+		</table>
+	{/if}
+</div>
+
+{#if feedbacks.length > 0}
+	<div class=" flex flex-col justify-end w-full text-right gap-1">
+		<div class="line-clamp-1 text-gray-500 text-xs">
+			{$i18n.t('Help us create the best community leaderboard by sharing your feedback history!')}
+		</div>
+
+		<div class="flex space-x-1 ml-auto">
+			<Tooltip
+				content={$i18n.t(
+					'To protect your privacy, only ratings, model IDs, tags, and metadata are shared from your feedback—your chat logs remain private and are not included.'
+				)}
+			>
+				<button
+					class="flex text-xs items-center px-3 py-1.5 rounded-xl bg-gray-50 hover:bg-gray-100 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-200 transition"
+					on:click={async () => {
+						shareHandler();
+					}}
+				>
+					<div class=" self-center mr-2 font-medium line-clamp-1">
+						{$i18n.t('Share to OpenWebUI Community')}
+					</div>
+
+					<div class=" self-center">
+						<CloudArrowUp className="size-3" strokeWidth="3" />
+					</div>
+				</button>
+			</Tooltip>
+		</div>
+	</div>
+{/if}
+
+{#if feedbacks.length > 10}
+	<Pagination bind:page count={feedbacks.length} perPage={10} />
+{/if}

src/lib/components/admin/Evaluations/Leaderboard.svelte

@@ -0,0 +1,410 @@
+<script lang="ts">
+	import * as ort from 'onnxruntime-web';
+	import { AutoModel, AutoTokenizer } from '@huggingface/transformers';
+
+	import { onMount, getContext } from 'svelte';
+	import { models } from '$lib/stores';
+
+	import Spinner from '$lib/components/common/Spinner.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import MagnifyingGlass from '$lib/components/icons/MagnifyingGlass.svelte';
+
+	const i18n = getContext('i18n');
+
+	const EMBEDDING_MODEL = 'TaylorAI/bge-micro-v2';
+
+	let tokenizer = null;
+	let model = null;
+
+	export let feedbacks = [];
+
+	let rankedModels = [];
+
+	let query = '';
+
+	let tagEmbeddings = new Map();
+	let loadingLeaderboard = true;
+	let debounceTimer;
+
+	type Feedback = {
+		id: string;
+		data: {
+			rating: number;
+			model_id: string;
+			sibling_model_ids: string[] | null;
+			reason: string;
+			comment: string;
+			tags: string[];
+		};
+		user: {
+			name: string;
+			profile_image_url: string;
+		};
+		updated_at: number;
+	};
+
+	type ModelStats = {
+		rating: number;
+		won: number;
+		lost: number;
+	};
+
+	//////////////////////
+	//
+	// Rank models by Elo rating
+	//
+	//////////////////////
+
+	const rankHandler = async (similarities: Map<string, number> = new Map()) => {
+		const modelStats = calculateModelStats(feedbacks, similarities);
+
+		rankedModels = $models
+			.filter((m) => m?.owned_by !== 'arena' && (m?.info?.meta?.hidden ?? false) !== true)
+			.map((model) => {
+				const stats = modelStats.get(model.id);
+				return {
+					...model,
+					rating: stats ? Math.round(stats.rating) : '-',
+					stats: {
+						count: stats ? stats.won + stats.lost : 0,
+						won: stats ? stats.won.toString() : '-',
+						lost: stats ? stats.lost.toString() : '-'
+					}
+				};
+			})
+			.sort((a, b) => {
+				if (a.rating === '-' && b.rating !== '-') return 1;
+				if (b.rating === '-' && a.rating !== '-') return -1;
+				if (a.rating !== '-' && b.rating !== '-') return b.rating - a.rating;
+				return a.name.localeCompare(b.name);
+			});
+
+		loadingLeaderboard = false;
+	};
+
+	function calculateModelStats(
+		feedbacks: Feedback[],
+		similarities: Map<string, number>
+	): Map<string, ModelStats> {
+		const stats = new Map<string, ModelStats>();
+		const K = 32;
+
+		function getOrDefaultStats(modelId: string): ModelStats {
+			return stats.get(modelId) || { rating: 1000, won: 0, lost: 0 };
+		}
+
+		function updateStats(modelId: string, ratingChange: number, outcome: number) {
+			const currentStats = getOrDefaultStats(modelId);
+			currentStats.rating += ratingChange;
+			if (outcome === 1) currentStats.won++;
+			else if (outcome === 0) currentStats.lost++;
+			stats.set(modelId, currentStats);
+		}
+
+		function calculateEloChange(
+			ratingA: number,
+			ratingB: number,
+			outcome: number,
+			similarity: number
+		): number {
+			const expectedScore = 1 / (1 + Math.pow(10, (ratingB - ratingA) / 400));
+			return K * (outcome - expectedScore) * similarity;
+		}
+
+		feedbacks.forEach((feedback) => {
+			const modelA = feedback.data.model_id;
+			const statsA = getOrDefaultStats(modelA);
+			let outcome: number;
+
+			switch (feedback.data.rating.toString()) {
+				case '1':
+					outcome = 1;
+					break;
+				case '-1':
+					outcome = 0;
+					break;
+				default:
+					return; // Skip invalid ratings
+			}
+
+			// If the query is empty, set similarity to 1, else get the similarity from the map
+			const similarity = query !== '' ? similarities.get(feedback.id) || 0 : 1;
+			const opponents = feedback.data.sibling_model_ids || [];
+
+			opponents.forEach((modelB) => {
+				const statsB = getOrDefaultStats(modelB);
+				const changeA = calculateEloChange(statsA.rating, statsB.rating, outcome, similarity);
+				const changeB = calculateEloChange(statsB.rating, statsA.rating, 1 - outcome, similarity);
+
+				updateStats(modelA, changeA, outcome);
+				updateStats(modelB, changeB, 1 - outcome);
+			});
+		});
+
+		return stats;
+	}
+
+	//////////////////////
+	//
+	// Calculate cosine similarity
+	//
+	//////////////////////
+
+	const cosineSimilarity = (vecA, vecB) => {
+		// Ensure the lengths of the vectors are the same
+		if (vecA.length !== vecB.length) {
+			throw new Error('Vectors must be the same length');
+		}
+
+		// Calculate the dot product
+		let dotProduct = 0;
+		let normA = 0;
+		let normB = 0;
+
+		for (let i = 0; i < vecA.length; i++) {
+			dotProduct += vecA[i] * vecB[i];
+			normA += vecA[i] ** 2;
+			normB += vecB[i] ** 2;
+		}
+
+		// Calculate the magnitudes
+		normA = Math.sqrt(normA);
+		normB = Math.sqrt(normB);
+
+		// Avoid division by zero
+		if (normA === 0 || normB === 0) {
+			return 0;
+		}
+
+		// Return the cosine similarity
+		return dotProduct / (normA * normB);
+	};
+
+	const calculateMaxSimilarity = (queryEmbedding, tagEmbeddings: Map<string, number[]>) => {
+		let maxSimilarity = 0;
+		for (const tagEmbedding of tagEmbeddings.values()) {
+			const similarity = cosineSimilarity(queryEmbedding, tagEmbedding);
+			maxSimilarity = Math.max(maxSimilarity, similarity);
+		}
+		return maxSimilarity;
+	};
+
+	//////////////////////
+	//
+	// Embedding functions
+	//
+	//////////////////////
+
+	const loadEmbeddingModel = async () => {
+		// Check if the tokenizer and model are already loaded and stored in the window object
+		if (!window.tokenizer) {
+			window.tokenizer = await AutoTokenizer.from_pretrained(EMBEDDING_MODEL);
+		}
+
+		if (!window.model) {
+			window.model = await AutoModel.from_pretrained(EMBEDDING_MODEL);
+		}
+
+		// Use the tokenizer and model from the window object
+		tokenizer = window.tokenizer;
+		model = window.model;
+
+		// Pre-compute embeddings for all unique tags
+		const allTags = new Set(feedbacks.flatMap((feedback) => feedback.data.tags || []));
+		await getTagEmbeddings(Array.from(allTags));
+	};
+
+	const getEmbeddings = async (text: string) => {
+		const tokens = await tokenizer(text);
+		const output = await model(tokens);
+
+		// Perform mean pooling on the last hidden states
+		const embeddings = output.last_hidden_state.mean(1);
+		return embeddings.ort_tensor.data;
+	};
+
+	const getTagEmbeddings = async (tags: string[]) => {
+		const embeddings = new Map();
+		for (const tag of tags) {
+			if (!tagEmbeddings.has(tag)) {
+				tagEmbeddings.set(tag, await getEmbeddings(tag));
+			}
+			embeddings.set(tag, tagEmbeddings.get(tag));
+		}
+		return embeddings;
+	};
+
+	const debouncedQueryHandler = async () => {
+		loadingLeaderboard = true;
+
+		if (query.trim() === '') {
+			rankHandler();
+			return;
+		}
+
+		clearTimeout(debounceTimer);
+
+		debounceTimer = setTimeout(async () => {
+			const queryEmbedding = await getEmbeddings(query);
+			const similarities = new Map<string, number>();
+
+			for (const feedback of feedbacks) {
+				const feedbackTags = feedback.data.tags || [];
+				const tagEmbeddings = await getTagEmbeddings(feedbackTags);
+				const maxSimilarity = calculateMaxSimilarity(queryEmbedding, tagEmbeddings);
+				similarities.set(feedback.id, maxSimilarity);
+			}
+
+			rankHandler(similarities);
+		}, 1500); // Debounce for 1.5 seconds
+	};
+
+	$: query, debouncedQueryHandler();
+
+	onMount(async () => {
+		rankHandler();
+	});
+</script>
+
+<div class="mt-0.5 mb-2 gap-1 flex flex-col md:flex-row justify-between">
+	<div class="flex md:self-center text-lg font-medium px-0.5 shrink-0 items-center">
+		<div class=" gap-1">
+			{$i18n.t('Leaderboard')}
+		</div>
+
+		<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
+
+		<span class="text-lg font-medium text-gray-500 dark:text-gray-300 mr-1.5"
+			>{rankedModels.length}</span
+		>
+	</div>
+
+	<div class=" flex space-x-2">
+		<Tooltip content={$i18n.t('Re-rank models by topic similarity')}>
+			<div class="flex flex-1">
+				<div class=" self-center ml-1 mr-3">
+					<MagnifyingGlass className="size-3" />
+				</div>
+				<input
+					class=" w-full text-sm pr-4 py-1 rounded-r-xl outline-none bg-transparent"
+					bind:value={query}
+					placeholder={$i18n.t('Search')}
+					on:focus={() => {
+						loadEmbeddingModel();
+					}}
+				/>
+			</div>
+		</Tooltip>
+	</div>
+</div>
+
+<div class="scrollbar-hidden relative whitespace-nowrap overflow-x-auto max-w-full rounded pt-0.5">
+	{#if loadingLeaderboard}
+		<div class=" absolute top-0 bottom-0 left-0 right-0 flex">
+			<div class="m-auto">
+				<Spinner />
+			</div>
+		</div>
+	{/if}
+	{#if (rankedModels ?? []).length === 0}
+		<div class="text-center text-xs text-gray-500 dark:text-gray-400 py-1">
+			{$i18n.t('No models found')}
+		</div>
+	{:else}
+		<table
+			class="w-full text-sm text-left text-gray-500 dark:text-gray-400 table-auto max-w-full rounded {loadingLeaderboard
+				? 'opacity-20'
+				: ''}"
+		>
+			<thead
+				class="text-xs text-gray-700 uppercase bg-gray-50 dark:bg-gray-850 dark:text-gray-400 -translate-y-0.5"
+			>
+				<tr class="">
+					<th scope="col" class="px-3 py-1.5 cursor-pointer select-none w-3">
+						{$i18n.t('RK')}
+					</th>
+					<th scope="col" class="px-3 py-1.5 cursor-pointer select-none">
+						{$i18n.t('Model')}
+					</th>
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-fit">
+						{$i18n.t('Rating')}
+					</th>
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-5">
+						{$i18n.t('Won')}
+					</th>
+					<th scope="col" class="px-3 py-1.5 text-right cursor-pointer select-none w-5">
+						{$i18n.t('Lost')}
+					</th>
+				</tr>
+			</thead>
+			<tbody class="">
+				{#each rankedModels as model, modelIdx (model.id)}
+					<tr class="bg-white dark:bg-gray-900 dark:border-gray-850 text-xs group">
+						<td class="px-3 py-1.5 text-left font-medium text-gray-900 dark:text-white w-fit">
+							<div class=" line-clamp-1">
+								{model?.rating !== '-' ? modelIdx + 1 : '-'}
+							</div>
+						</td>
+						<td class="px-3 py-1.5 flex flex-col justify-center">
+							<div class="flex items-center gap-2">
+								<div class="flex-shrink-0">
+									<img
+										src={model?.info?.meta?.profile_image_url ?? '/favicon.png'}
+										alt={model.name}
+										class="size-5 rounded-full object-cover shrink-0"
+									/>
+								</div>
+
+								<div class="font-medium text-gray-800 dark:text-gray-200 pr-4">
+									{model.name}
+								</div>
+							</div>
+						</td>
+						<td class="px-3 py-1.5 text-right font-medium text-gray-900 dark:text-white w-max">
+							{model.rating}
+						</td>
+
+						<td class=" px-3 py-1.5 text-right font-semibold text-green-500">
+							<div class=" w-10">
+								{#if model.stats.won === '-'}
+									-
+								{:else}
+									<span class="hidden group-hover:inline"
+										>{((model.stats.won / model.stats.count) * 100).toFixed(1)}%</span
+									>
+									<span class=" group-hover:hidden">{model.stats.won}</span>
+								{/if}
+							</div>
+						</td>
+
+						<td class="px-3 py-1.5 text-right font-semibold text-red-500">
+							<div class=" w-10">
+								{#if model.stats.lost === '-'}
+									-
+								{:else}
+									<span class="hidden group-hover:inline"
+										>{((model.stats.lost / model.stats.count) * 100).toFixed(1)}%</span
+									>
+									<span class=" group-hover:hidden">{model.stats.lost}</span>
+								{/if}
+							</div>
+						</td>
+					</tr>
+				{/each}
+			</tbody>
+		</table>
+	{/if}
+</div>
+
+<div class=" text-gray-500 text-xs mt-1.5 w-full flex justify-end">
+	<div class=" text-right">
+		<div class="line-clamp-1">
+			ⓘ {$i18n.t(
+				'The evaluation leaderboard is based on the Elo rating system and is updated in real-time.'
+			)}
+		</div>
+		{$i18n.t(
+			'The leaderboard is currently in beta, and we may adjust the rating calculations as we refine the algorithm.'
+		)}
+	</div>
+</div>

src/lib/components/workspace/Functions.svelte → src/lib/components/admin/Functions.svelte

@@ -5,7 +5,6 @@
 
 	import { WEBUI_NAME, config, functions, models } from '$lib/stores';
 	import { onMount, getContext, tick } from 'svelte';
-	import { createNewPrompt, deletePromptByCommand, getPrompts } from '$lib/apis/prompts';
 
 	import { goto } from '$app/navigation';
 	import {
@@ -25,11 +24,14 @@
 	import FunctionMenu from './Functions/FunctionMenu.svelte';
 	import EllipsisHorizontal from '../icons/EllipsisHorizontal.svelte';
 	import Switch from '../common/Switch.svelte';
-	import ValvesModal from './common/ValvesModal.svelte';
-	import ManifestModal from './common/ManifestModal.svelte';
+	import ValvesModal from '../workspace/common/ValvesModal.svelte';
+	import ManifestModal from '../workspace/common/ManifestModal.svelte';
 	import Heart from '../icons/Heart.svelte';
 	import DeleteConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
 	import GarbageBin from '../icons/GarbageBin.svelte';
+	import Search from '../icons/Search.svelte';
+	import Plus from '../icons/Plus.svelte';
+	import ChevronRight from '../icons/ChevronRight.svelte';
 
 	const i18n = getContext('i18n');
 
@@ -48,12 +50,14 @@
 	let showDeleteConfirm = false;
 
 	let filteredItems = [];
-	$: filteredItems = $functions.filter(
-		(f) =>
-			query === '' ||
-			f.name.toLowerCase().includes(query.toLowerCase()) ||
-			f.id.toLowerCase().includes(query.toLowerCase())
-	);
+	$: filteredItems = $functions
+		.filter(
+			(f) =>
+				query === '' ||
+				f.name.toLowerCase().includes(query.toLowerCase()) ||
+				f.id.toLowerCase().includes(query.toLowerCase())
+		)
+		.sort((a, b) => a.type.localeCompare(b.type) || a.name.localeCompare(b.name));
 
 	const shareHandler = async (func) => {
 		const item = await getFunctionById(localStorage.token, func.id).catch((error) => {
@@ -94,7 +98,7 @@
 				id: `${_function.id}_clone`,
 				name: `${_function.name} (Clone)`
 			});
-			goto('/workspace/functions/create');
+			goto('/admin/functions/create');
 		}
 	};
 
@@ -182,68 +186,46 @@
 	</title>
 </svelte:head>
 
-<div class=" flex w-full space-x-2 mb-2.5">
-	<div class="flex flex-1">
-		<div class=" self-center ml-1 mr-3">
-			<svg
-				xmlns="http://www.w3.org/2000/svg"
-				viewBox="0 0 20 20"
-				fill="currentColor"
-				class="w-4 h-4"
-			>
-				<path
-					fill-rule="evenodd"
-					d="M9 3.5a5.5 5.5 0 100 11 5.5 5.5 0 000-11zM2 9a7 7 0 1112.452 4.391l3.328 3.329a.75.75 0 11-1.06 1.06l-3.329-3.328A7 7 0 012 9z"
-					clip-rule="evenodd"
-				/>
-			</svg>
+<div class="flex flex-col gap-1 mt-1.5 mb-2">
+	<div class="flex justify-between items-center">
+		<div class="flex md:self-center text-xl items-center font-medium px-0.5">
+			{$i18n.t('Functions')}
+			<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
+			<span class="text-base font-lg text-gray-500 dark:text-gray-300">{filteredItems.length}</span>
 		</div>
-		<input
-			class=" w-full text-sm pr-4 py-1 rounded-r-xl outline-none bg-transparent"
-			bind:value={query}
-			placeholder={$i18n.t('Search Functions')}
-		/>
 	</div>
 
-	<div>
-		<a
-			class=" px-2 py-2 rounded-xl border border-gray-200 dark:border-gray-600 dark:border-0 hover:bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 transition font-medium text-sm flex items-center space-x-1"
-			href="/workspace/functions/create"
-		>
-			<svg
-				xmlns="http://www.w3.org/2000/svg"
-				viewBox="0 0 16 16"
-				fill="currentColor"
-				class="w-4 h-4"
-			>
-				<path
-					d="M8.75 3.75a.75.75 0 0 0-1.5 0v3.5h-3.5a.75.75 0 0 0 0 1.5h3.5v3.5a.75.75 0 0 0 1.5 0v-3.5h3.5a.75.75 0 0 0 0-1.5h-3.5v-3.5Z"
-				/>
-			</svg>
-		</a>
-	</div>
-</div>
+	<div class=" flex w-full space-x-2">
+		<div class="flex flex-1">
+			<div class=" self-center ml-1 mr-3">
+				<Search className="size-3.5" />
+			</div>
+			<input
+				class=" w-full text-sm pr-4 py-1 rounded-r-xl outline-none bg-transparent"
+				bind:value={query}
+				placeholder={$i18n.t('Search Functions')}
+			/>
+		</div>
 
-<div class="mb-3.5">
-	<div class="flex justify-between items-center">
-		<div class="flex md:self-center text-base font-medium px-0.5">
-			{$i18n.t('Functions')}
-			<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
-			<span class="text-base font-medium text-gray-500 dark:text-gray-300"
-				>{filteredItems.length}</span
+		<div>
+			<a
+				class=" px-2 py-2 rounded-xl hover:bg-gray-700/10 dark:hover:bg-gray-100/10 dark:text-gray-300 dark:hover:text-white transition font-medium text-sm flex items-center space-x-1"
+				href="/admin/functions/create"
 			>
+				<Plus className="size-3.5" />
+			</a>
 		</div>
 	</div>
 </div>
 
-<div class="my-3 mb-5">
+<div class="mb-5">
 	{#each filteredItems as func}
 		<div
 			class=" flex space-x-4 cursor-pointer w-full px-3 py-2 dark:hover:bg-white/5 hover:bg-black/5 rounded-xl"
 		>
 			<a
 				class=" flex flex-1 space-x-3.5 cursor-pointer w-full"
-				href={`/workspace/functions/edit?id=${encodeURIComponent(func.id)}`}
+				href={`/admin/functions/edit?id=${encodeURIComponent(func.id)}`}
 			>
 				<div class="flex items-center text-left">
 					<div class=" flex-1 self-center pl-1">
@@ -340,7 +322,7 @@
 					<FunctionMenu
 						{func}
 						editHandler={() => {
-							goto(`/workspace/functions/edit?id=${encodeURIComponent(func.id)}`);
+							goto(`/admin/functions/edit?id=${encodeURIComponent(func.id)}`);
 						}}
 						shareHandler={() => {
 							shareHandler(func);
@@ -470,40 +452,27 @@
 
 {#if $config?.features.enable_community_sharing}
 	<div class=" my-16">
-		<div class=" text-lg font-semibold mb-3 line-clamp-1">
+		<div class=" text-xl font-medium mb-1 line-clamp-1">
 			{$i18n.t('Made by OpenWebUI Community')}
 		</div>
 
 		<a
-			class=" flex space-x-4 cursor-pointer w-full mb-2 px-3 py-2"
+			class=" flex cursor-pointer items-center justify-between hover:bg-gray-50 dark:hover:bg-gray-850 w-full mb-2 px-3.5 py-1.5 rounded-xl transition"
 			href="https://openwebui.com/#open-webui-community"
 			target="_blank"
 		>
-			<div class=" self-center w-10 flex-shrink-0">
-				<div
-					class="w-full h-10 flex justify-center rounded-full bg-transparent dark:bg-gray-700 border border-dashed border-gray-200"
-				>
-					<svg
-						xmlns="http://www.w3.org/2000/svg"
-						viewBox="0 0 24 24"
-						fill="currentColor"
-						class="w-6"
-					>
-						<path
-							fill-rule="evenodd"
-							d="M12 3.75a.75.75 0 01.75.75v6.75h6.75a.75.75 0 010 1.5h-6.75v6.75a.75.75 0 01-1.5 0v-6.75H4.5a.75.75 0 010-1.5h6.75V4.5a.75.75 0 01.75-.75z"
-							clip-rule="evenodd"
-						/>
-					</svg>
-				</div>
-			</div>
-
 			<div class=" self-center">
 				<div class=" font-semibold line-clamp-1">{$i18n.t('Discover a function')}</div>
 				<div class=" text-sm line-clamp-1">
 					{$i18n.t('Discover, download, and explore custom functions')}
 				</div>
 			</div>
+
+			<div>
+				<div>
+					<ChevronRight />
+				</div>
+			</div>
 		</a>
 	</div>
 {/if}

src/lib/components/workspace/Functions/FunctionEditor.svelte → src/lib/components/admin/Functions/FunctionEditor.svelte

@@ -305,7 +305,7 @@ class Pipe:
 								<button
 									class="w-full text-left text-sm py-1.5 px-1 rounded-lg dark:text-gray-300 dark:hover:text-white hover:bg-black/5 dark:hover:bg-gray-850"
 									on:click={() => {
-										goto('/workspace/functions');
+										goto('/admin/functions');
 									}}
 									type="button"
 								>
@@ -315,13 +315,15 @@ class Pipe:
 						</div>
 
 						<div class="flex-1">
-							<input
-								class="w-full text-2xl font-medium bg-transparent outline-none font-primary"
-								type="text"
-								placeholder={$i18n.t('Function Name (e.g. My Filter)')}
-								bind:value={name}
-								required
-							/>
+							<Tooltip content={$i18n.t('e.g. My Filter')} placement="top-start">
+								<input
+									class="w-full text-2xl font-medium bg-transparent outline-none font-primary"
+									type="text"
+									placeholder={$i18n.t('Function Name')}
+									bind:value={name}
+									required
+								/>
+							</Tooltip>
 						</div>
 
 						<div>
@@ -329,31 +331,37 @@ class Pipe:
 						</div>
 					</div>
 
-					<div class=" flex gap-2 px-1">
+					<div class=" flex gap-2 px-1 items-center">
 						{#if edit}
 							<div class="text-sm text-gray-500 flex-shrink-0">
 								{id}
 							</div>
 						{:else}
+							<Tooltip className="w-full" content={$i18n.t('e.g. my_filter')} placement="top-start">
+								<input
+									class="w-full text-sm disabled:text-gray-500 bg-transparent outline-none"
+									type="text"
+									placeholder={$i18n.t('Function ID')}
+									bind:value={id}
+									required
+									disabled={edit}
+								/>
+							</Tooltip>
+						{/if}
+
+						<Tooltip
+							className="w-full self-center items-center flex"
+							content={$i18n.t('e.g. A filter to remove profanity from text')}
+							placement="top-start"
+						>
 							<input
-								class="w-full text-sm disabled:text-gray-500 bg-transparent outline-none"
+								class="w-full text-sm bg-transparent outline-none"
 								type="text"
-								placeholder={$i18n.t('Function ID (e.g. my_filter)')}
-								bind:value={id}
+								placeholder={$i18n.t('Function Description')}
+								bind:value={meta.description}
 								required
-								disabled={edit}
 							/>
-						{/if}
-
-						<input
-							class="w-full text-sm bg-transparent outline-none"
-							type="text"
-							placeholder={$i18n.t(
-								'Function Description (e.g. A filter to remove profanity from text)'
-							)}
-							bind:value={meta.description}
-							required
-						/>
+						</Tooltip>
 					</div>
 				</div>
 

src/lib/components/admin/Settings.svelte

@@ -2,11 +2,11 @@
 	import { getContext, tick, onMount } from 'svelte';
 	import { toast } from 'svelte-sonner';
 
+	import { config } from '$lib/stores';
+	import { getBackendConfig } from '$lib/apis';
 	import Database from './Settings/Database.svelte';
 
 	import General from './Settings/General.svelte';
-	import Users from './Settings/Users.svelte';
-
 	import Pipelines from './Settings/Pipelines.svelte';
 	import Audio from './Settings/Audio.svelte';
 	import Images from './Settings/Images.svelte';
@@ -15,8 +15,7 @@
 	import Connections from './Settings/Connections.svelte';
 	import Documents from './Settings/Documents.svelte';
 	import WebSearch from './Settings/WebSearch.svelte';
-	import { config } from '$lib/stores';
-	import { getBackendConfig } from '$lib/apis';
+
 	import ChartBar from '../icons/ChartBar.svelte';
 	import DocumentChartBar from '../icons/DocumentChartBar.svelte';
 	import Evaluations from './Settings/Evaluations.svelte';
@@ -39,16 +38,16 @@
 	});
 </script>
 
-<div class="flex flex-col lg:flex-row w-full h-full py-2 lg:space-x-4">
+<div class="flex flex-col lg:flex-row w-full h-full pb-2 lg:space-x-4">
 	<div
 		id="admin-settings-tabs-container"
-		class="tabs flex flex-row overflow-x-auto space-x-1 max-w-full lg:space-x-0 lg:space-y-1 lg:flex-col lg:flex-none lg:w-44 dark:text-gray-200 text-xs text-left scrollbar-none"
+		class="tabs flex flex-row overflow-x-auto gap-2.5 max-w-full lg:gap-1 lg:flex-col lg:flex-none lg:w-40 dark:text-gray-200 text-sm font-medium text-left scrollbar-none"
 	>
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 lg:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 lg:flex-none flex text-right transition {selectedTab ===
 			'general'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'general';
 			}}
@@ -71,34 +70,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
-			'users'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
-			on:click={() => {
-				selectedTab = 'users';
-			}}
-		>
-			<div class=" self-center mr-2">
-				<svg
-					xmlns="http://www.w3.org/2000/svg"
-					viewBox="0 0 16 16"
-					fill="currentColor"
-					class="w-4 h-4"
-				>
-					<path
-						d="M8 8a2.5 2.5 0 1 0 0-5 2.5 2.5 0 0 0 0 5ZM3.156 11.763c.16-.629.44-1.21.813-1.72a2.5 2.5 0 0 0-2.725 1.377c-.136.287.102.58.418.58h1.449c.01-.077.025-.156.045-.237ZM12.847 11.763c.02.08.036.16.046.237h1.446c.316 0 .554-.293.417-.579a2.5 2.5 0 0 0-2.722-1.378c.374.51.653 1.09.813 1.72ZM14 7.5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0ZM3.5 9a1.5 1.5 0 1 0 0-3 1.5 1.5 0 0 0 0 3ZM5 13c-.552 0-1.013-.455-.876-.99a4.002 4.002 0 0 1 7.753 0c.136.535-.324.99-.877.99H5Z"
-					/>
-				</svg>
-			</div>
-			<div class=" self-center">{$i18n.t('Users')}</div>
-		</button>
-
-		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'connections'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'connections';
 			}}
@@ -119,10 +94,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'models'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'models';
 			}}
@@ -145,10 +120,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'evaluations'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'evaluations';
 			}}
@@ -160,10 +135,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'documents'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'documents';
 			}}
@@ -190,10 +165,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'web'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'web';
 			}}
@@ -214,10 +189,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'interface'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'interface';
 			}}
@@ -240,10 +215,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'audio'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'audio';
 			}}
@@ -267,10 +242,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'images'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'images';
 			}}
@@ -293,10 +268,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'pipelines'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'pipelines';
 			}}
@@ -323,10 +298,10 @@
 		</button>
 
 		<button
-			class="px-2.5 py-2 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
+			class="px-0.5 py-1 min-w-fit rounded-lg flex-1 md:flex-none flex text-right transition {selectedTab ===
 			'db'
-				? 'bg-gray-100 dark:bg-gray-800'
-				: ' hover:bg-gray-50 dark:hover:bg-gray-850'}"
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
 			on:click={() => {
 				selectedTab = 'db';
 			}}
@@ -351,7 +326,7 @@
 		</button>
 	</div>
 
-	<div class="flex-1 mt-3 lg:mt-0 overflow-y-scroll">
+	<div class="flex-1 mt-3 lg:mt-0 overflow-y-scroll pr-1 scrollbar-hidden">
 		{#if selectedTab === 'general'}
 			<General
 				saveHandler={async () => {
@@ -361,12 +336,6 @@
 					await config.set(await getBackendConfig());
 				}}
 			/>
-		{:else if selectedTab === 'users'}
-			<Users
-				saveHandler={() => {
-					toast.success($i18n.t('Settings saved successfully!'));
-				}}
-			/>
 		{:else if selectedTab === 'connections'}
 			<Connections
 				on:save={() => {

src/lib/components/admin/Settings/Audio.svelte

@@ -181,7 +181,7 @@
 					<div>
 						<div class="mt-1 flex gap-2 mb-1">
 							<input
-								class="flex-1 w-full rounded-lg py-2 pl-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+								class="flex-1 w-full bg-transparent outline-none"
 								placeholder={$i18n.t('API Base URL')}
 								bind:value={STT_OPENAI_API_BASE_URL}
 								required
@@ -322,6 +322,7 @@
 							}}
 						>
 							<option value="">{$i18n.t('Web API')}</option>
+							<option value="transformers">{$i18n.t('Transformers')} ({$i18n.t('Local')})</option>
 							<option value="openai">{$i18n.t('OpenAI')}</option>
 							<option value="elevenlabs">{$i18n.t('ElevenLabs')}</option>
 							<option value="azure">{$i18n.t('Azure AI Speech')}</option>
@@ -333,7 +334,7 @@
 					<div>
 						<div class="mt-1 flex gap-2 mb-1">
 							<input
-								class="flex-1 w-full rounded-lg py-2 pl-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+								class="flex-1 w-full bg-transparent outline-none"
 								placeholder={$i18n.t('API Base URL')}
 								bind:value={TTS_OPENAI_API_BASE_URL}
 								required
@@ -396,6 +397,47 @@
 							</div>
 						</div>
 					</div>
+				{:else if TTS_ENGINE === 'transformers'}
+					<div>
+						<div class=" mb-1.5 text-sm font-medium">{$i18n.t('TTS Model')}</div>
+						<div class="flex w-full">
+							<div class="flex-1">
+								<input
+									list="model-list"
+									class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+									bind:value={TTS_MODEL}
+									placeholder="CMU ARCTIC speaker embedding name"
+								/>
+
+								<datalist id="model-list">
+									<option value="tts-1" />
+								</datalist>
+							</div>
+						</div>
+						<div class="mt-2 mb-1 text-xs text-gray-400 dark:text-gray-500">
+							{$i18n.t(`Open WebUI uses SpeechT5 and CMU Arctic speaker embeddings.`)}
+
+							To learn more about SpeechT5,
+
+							<a
+								class=" hover:underline dark:text-gray-200 text-gray-800"
+								href="https://github.com/microsoft/SpeechT5"
+								target="_blank"
+							>
+								{$i18n.t(`click here`, {
+									name: 'SpeechT5'
+								})}.
+							</a>
+							To see the available CMU Arctic speaker embeddings,
+							<a
+								class=" hover:underline dark:text-gray-200 text-gray-800"
+								href="https://huggingface.co/datasets/Matthijs/cmu-arctic-xvectors"
+								target="_blank"
+							>
+								{$i18n.t(`click here`)}.
+							</a>
+						</div>
+					</div>
 				{:else if TTS_ENGINE === 'openai'}
 					<div class=" flex gap-2">
 						<div class="w-full">

src/lib/components/admin/Settings/Connections.svelte

@@ -1,31 +1,23 @@
 <script lang="ts">
-	import { models, user } from '$lib/stores';
+	import { toast } from 'svelte-sonner';
 	import { createEventDispatcher, onMount, getContext, tick } from 'svelte';
 
 	const dispatch = createEventDispatcher();
 
-	import {
-		getOllamaConfig,
-		getOllamaUrls,
-		getOllamaVersion,
-		updateOllamaConfig,
-		updateOllamaUrls
-	} from '$lib/apis/ollama';
-	import {
-		getOpenAIConfig,
-		getOpenAIKeys,
-		getOpenAIModels,
-		getOpenAIUrls,
-		updateOpenAIConfig,
-		updateOpenAIKeys,
-		updateOpenAIUrls
-	} from '$lib/apis/openai';
-	import { toast } from 'svelte-sonner';
+	import { getOllamaConfig, updateOllamaConfig } from '$lib/apis/ollama';
+	import { getOpenAIConfig, updateOpenAIConfig, getOpenAIModels } from '$lib/apis/openai';
+	import { getModels as _getModels } from '$lib/apis';
+
+	import { models, user } from '$lib/stores';
+
 	import Switch from '$lib/components/common/Switch.svelte';
 	import Spinner from '$lib/components/common/Spinner.svelte';
 	import Tooltip from '$lib/components/common/Tooltip.svelte';
-	import { getModels as _getModels } from '$lib/apis';
-	import SensitiveInput from '$lib/components/common/SensitiveInput.svelte';
+	import Plus from '$lib/components/icons/Plus.svelte';
+
+	import OpenAIConnection from './Connections/OpenAIConnection.svelte';
+	import AddConnectionModal from './Connections/AddConnectionModal.svelte';
+	import OllamaConnection from './Connections/OllamaConnection.svelte';
 
 	const i18n = getContext('i18n');
 
@@ -36,395 +28,302 @@
 
 	// External
 	let OLLAMA_BASE_URLS = [''];
+	let OLLAMA_API_CONFIGS = {};
 
 	let OPENAI_API_KEYS = [''];
 	let OPENAI_API_BASE_URLS = [''];
+	let OPENAI_API_CONFIGS = {};
 
-	let pipelineUrls = {};
+	let ENABLE_OPENAI_API: null | boolean = null;
+	let ENABLE_OLLAMA_API: null | boolean = null;
 
-	let ENABLE_OPENAI_API = null;
-	let ENABLE_OLLAMA_API = null;
+	let pipelineUrls = {};
+	let showAddOpenAIConnectionModal = false;
+	let showAddOllamaConnectionModal = false;
 
-	const verifyOpenAIHandler = async (idx) => {
-		OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS.map((url) => url.replace(/\/$/, ''));
+	const updateOpenAIHandler = async () => {
+		if (ENABLE_OPENAI_API !== null) {
+			OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS.filter(
+				(url, urlIdx) => OPENAI_API_BASE_URLS.indexOf(url) === urlIdx && url !== ''
+			).map((url) => url.replace(/\/$/, ''));
+
+			// Check if API KEYS length is same than API URLS length
+			if (OPENAI_API_KEYS.length !== OPENAI_API_BASE_URLS.length) {
+				// if there are more keys than urls, remove the extra keys
+				if (OPENAI_API_KEYS.length > OPENAI_API_BASE_URLS.length) {
+					OPENAI_API_KEYS = OPENAI_API_KEYS.slice(0, OPENAI_API_BASE_URLS.length);
+				}
 
-		OPENAI_API_BASE_URLS = await updateOpenAIUrls(localStorage.token, OPENAI_API_BASE_URLS);
-		OPENAI_API_KEYS = await updateOpenAIKeys(localStorage.token, OPENAI_API_KEYS);
+				// if there are more urls than keys, add empty keys
+				if (OPENAI_API_KEYS.length < OPENAI_API_BASE_URLS.length) {
+					const diff = OPENAI_API_BASE_URLS.length - OPENAI_API_KEYS.length;
+					for (let i = 0; i < diff; i++) {
+						OPENAI_API_KEYS.push('');
+					}
+				}
+			}
 
-		const res = await getOpenAIModels(localStorage.token, idx).catch((error) => {
-			toast.error(error);
-			return null;
-		});
+			const res = await updateOpenAIConfig(localStorage.token, {
+				ENABLE_OPENAI_API: ENABLE_OPENAI_API,
+				OPENAI_API_BASE_URLS: OPENAI_API_BASE_URLS,
+				OPENAI_API_KEYS: OPENAI_API_KEYS,
+				OPENAI_API_CONFIGS: OPENAI_API_CONFIGS
+			}).catch((error) => {
+				toast.error(error);
+			});
 
-		if (res) {
-			toast.success($i18n.t('Server connection verified'));
-			if (res.pipelines) {
-				pipelineUrls[OPENAI_API_BASE_URLS[idx]] = true;
+			if (res) {
+				toast.success($i18n.t('OpenAI API settings updated'));
+				await models.set(await getModels());
 			}
 		}
-
-		await models.set(await getModels());
 	};
 
-	const verifyOllamaHandler = async (idx) => {
-		OLLAMA_BASE_URLS = OLLAMA_BASE_URLS.filter((url) => url !== '').map((url) =>
-			url.replace(/\/$/, '')
-		);
-
-		OLLAMA_BASE_URLS = await updateOllamaUrls(localStorage.token, OLLAMA_BASE_URLS);
-
-		const res = await getOllamaVersion(localStorage.token, idx).catch((error) => {
-			toast.error(error);
-			return null;
-		});
-
-		if (res) {
-			toast.success($i18n.t('Server connection verified'));
-		}
-
-		await models.set(await getModels());
-	};
+	const updateOllamaHandler = async () => {
+		if (ENABLE_OLLAMA_API !== null) {
+			// Remove duplicate URLs
+			OLLAMA_BASE_URLS = OLLAMA_BASE_URLS.filter(
+				(url, urlIdx) => OLLAMA_BASE_URLS.indexOf(url) === urlIdx && url !== ''
+			).map((url) => url.replace(/\/$/, ''));
 
-	const updateOpenAIHandler = async () => {
-		OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS.map((url) => url.replace(/\/$/, ''));
+			console.log(OLLAMA_BASE_URLS);
 
-		// Check if API KEYS length is same than API URLS length
-		if (OPENAI_API_KEYS.length !== OPENAI_API_BASE_URLS.length) {
-			// if there are more keys than urls, remove the extra keys
-			if (OPENAI_API_KEYS.length > OPENAI_API_BASE_URLS.length) {
-				OPENAI_API_KEYS = OPENAI_API_KEYS.slice(0, OPENAI_API_BASE_URLS.length);
+			if (OLLAMA_BASE_URLS.length === 0) {
+				ENABLE_OLLAMA_API = false;
+				toast.info($i18n.t('Ollama API disabled'));
 			}
 
-			// if there are more urls than keys, add empty keys
-			if (OPENAI_API_KEYS.length < OPENAI_API_BASE_URLS.length) {
-				const diff = OPENAI_API_BASE_URLS.length - OPENAI_API_KEYS.length;
-				for (let i = 0; i < diff; i++) {
-					OPENAI_API_KEYS.push('');
-				}
+			const res = await updateOllamaConfig(localStorage.token, {
+				ENABLE_OLLAMA_API: ENABLE_OLLAMA_API,
+				OLLAMA_BASE_URLS: OLLAMA_BASE_URLS,
+				OLLAMA_API_CONFIGS: OLLAMA_API_CONFIGS
+			}).catch((error) => {
+				toast.error(error);
+			});
+
+			if (res) {
+				toast.success($i18n.t('Ollama API settings updated'));
+				await models.set(await getModels());
 			}
 		}
-
-		OPENAI_API_BASE_URLS = await updateOpenAIUrls(localStorage.token, OPENAI_API_BASE_URLS);
-		OPENAI_API_KEYS = await updateOpenAIKeys(localStorage.token, OPENAI_API_KEYS);
-		await models.set(await getModels());
 	};
 
-	const updateOllamaUrlsHandler = async () => {
-		OLLAMA_BASE_URLS = OLLAMA_BASE_URLS.filter((url) => url !== '').map((url) =>
-			url.replace(/\/$/, '')
-		);
-
-		console.log(OLLAMA_BASE_URLS);
+	const addOpenAIConnectionHandler = async (connection) => {
+		OPENAI_API_BASE_URLS = [...OPENAI_API_BASE_URLS, connection.url];
+		OPENAI_API_KEYS = [...OPENAI_API_KEYS, connection.key];
+		OPENAI_API_CONFIGS[connection.url] = connection.config;
 
-		if (OLLAMA_BASE_URLS.length === 0) {
-			ENABLE_OLLAMA_API = false;
-			await updateOllamaConfig(localStorage.token, ENABLE_OLLAMA_API);
-
-			toast.info($i18n.t('Ollama API disabled'));
-		} else {
-			OLLAMA_BASE_URLS = await updateOllamaUrls(localStorage.token, OLLAMA_BASE_URLS);
+		await updateOpenAIHandler();
+	};
 
-			const ollamaVersion = await getOllamaVersion(localStorage.token).catch((error) => {
-				toast.error(error);
-				return null;
-			});
+	const addOllamaConnectionHandler = async (connection) => {
+		OLLAMA_BASE_URLS = [...OLLAMA_BASE_URLS, connection.url];
+		OLLAMA_API_CONFIGS[connection.url] = connection.config;
 
-			if (ollamaVersion) {
-				toast.success($i18n.t('Server connection verified'));
-				await models.set(await getModels());
-			}
-		}
+		await updateOllamaHandler();
 	};
 
 	onMount(async () => {
 		if ($user.role === 'admin') {
+			let ollamaConfig = {};
+			let openaiConfig = {};
+
 			await Promise.all([
 				(async () => {
-					OLLAMA_BASE_URLS = await getOllamaUrls(localStorage.token);
-				})(),
-				(async () => {
-					OPENAI_API_BASE_URLS = await getOpenAIUrls(localStorage.token);
+					ollamaConfig = await getOllamaConfig(localStorage.token);
 				})(),
 				(async () => {
-					OPENAI_API_KEYS = await getOpenAIKeys(localStorage.token);
+					openaiConfig = await getOpenAIConfig(localStorage.token);
 				})()
 			]);
 
-			const ollamaConfig = await getOllamaConfig(localStorage.token);
-			const openaiConfig = await getOpenAIConfig(localStorage.token);
-
 			ENABLE_OPENAI_API = openaiConfig.ENABLE_OPENAI_API;
 			ENABLE_OLLAMA_API = ollamaConfig.ENABLE_OLLAMA_API;
 
+			OPENAI_API_BASE_URLS = openaiConfig.OPENAI_API_BASE_URLS;
+			OPENAI_API_KEYS = openaiConfig.OPENAI_API_KEYS;
+			OPENAI_API_CONFIGS = openaiConfig.OPENAI_API_CONFIGS;
+
+			OLLAMA_BASE_URLS = ollamaConfig.OLLAMA_BASE_URLS;
+			OLLAMA_API_CONFIGS = ollamaConfig.OLLAMA_API_CONFIGS;
+
 			if (ENABLE_OPENAI_API) {
+				for (const url of OPENAI_API_BASE_URLS) {
+					if (!OPENAI_API_CONFIGS[url]) {
+						OPENAI_API_CONFIGS[url] = {};
+					}
+				}
+
 				OPENAI_API_BASE_URLS.forEach(async (url, idx) => {
+					OPENAI_API_CONFIGS[url] = OPENAI_API_CONFIGS[url] || {};
+					if (!(OPENAI_API_CONFIGS[url]?.enable ?? true)) {
+						return;
+					}
 					const res = await getOpenAIModels(localStorage.token, idx);
 					if (res.pipelines) {
 						pipelineUrls[url] = true;
 					}
 				});
 			}
+
+			if (ENABLE_OLLAMA_API) {
+				for (const url of OLLAMA_BASE_URLS) {
+					if (!OLLAMA_API_CONFIGS[url]) {
+						OLLAMA_API_CONFIGS[url] = {};
+					}
+				}
+			}
 		}
 	});
 </script>
 
+<AddConnectionModal
+	bind:show={showAddOpenAIConnectionModal}
+	onSubmit={addOpenAIConnectionHandler}
+/>
+
+<AddConnectionModal
+	ollama
+	bind:show={showAddOllamaConnectionModal}
+	onSubmit={addOllamaConnectionHandler}
+/>
+
 <form
 	class="flex flex-col h-full justify-between text-sm"
 	on:submit|preventDefault={() => {
 		updateOpenAIHandler();
-		updateOllamaUrlsHandler();
+		updateOllamaHandler();
 
 		dispatch('save');
 	}}
 >
-	<div class="space-y-3 overflow-y-scroll scrollbar-hidden h-full">
+	<div class=" overflow-y-scroll scrollbar-hidden h-full">
 		{#if ENABLE_OPENAI_API !== null && ENABLE_OLLAMA_API !== null}
-			<div class=" space-y-3">
+			<div class="my-2">
 				<div class="mt-2 space-y-2 pr-1.5">
 					<div class="flex justify-between items-center text-sm">
 						<div class="  font-medium">{$i18n.t('OpenAI API')}</div>
 
-						<div class="mt-1">
-							<Switch
-								bind:state={ENABLE_OPENAI_API}
-								on:change={async () => {
-									updateOpenAIConfig(localStorage.token, ENABLE_OPENAI_API);
-								}}
-							/>
+						<div class="flex items-center">
+							<div class="">
+								<Switch
+									bind:state={ENABLE_OPENAI_API}
+									on:change={async () => {
+										updateOpenAIHandler();
+									}}
+								/>
+							</div>
 						</div>
 					</div>
 
 					{#if ENABLE_OPENAI_API}
-						<div class="flex flex-col gap-1">
-							{#each OPENAI_API_BASE_URLS as url, idx}
-								<div class="flex w-full gap-2">
-									<div class="flex-1 relative">
-										<input
-											class="w-full rounded-lg py-2 px-4 {pipelineUrls[url]
-												? 'pr-8'
-												: ''} text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-											placeholder={$i18n.t('API Base URL')}
-											bind:value={url}
-											autocomplete="off"
-										/>
-
-										{#if pipelineUrls[url]}
-											<div class=" absolute top-2.5 right-2.5">
-												<Tooltip content="Pipelines">
-													<svg
-														xmlns="http://www.w3.org/2000/svg"
-														viewBox="0 0 24 24"
-														fill="currentColor"
-														class="size-4"
-													>
-														<path
-															d="M11.644 1.59a.75.75 0 0 1 .712 0l9.75 5.25a.75.75 0 0 1 0 1.32l-9.75 5.25a.75.75 0 0 1-.712 0l-9.75-5.25a.75.75 0 0 1 0-1.32l9.75-5.25Z"
-														/>
-														<path
-															d="m3.265 10.602 7.668 4.129a2.25 2.25 0 0 0 2.134 0l7.668-4.13 1.37.739a.75.75 0 0 1 0 1.32l-9.75 5.25a.75.75 0 0 1-.71 0l-9.75-5.25a.75.75 0 0 1 0-1.32l1.37-.738Z"
-														/>
-														<path
-															d="m10.933 19.231-7.668-4.13-1.37.739a.75.75 0 0 0 0 1.32l9.75 5.25c.221.12.489.12.71 0l9.75-5.25a.75.75 0 0 0 0-1.32l-1.37-.738-7.668 4.13a2.25 2.25 0 0 1-2.134-.001Z"
-														/>
-													</svg>
-												</Tooltip>
-											</div>
-										{/if}
-									</div>
-
-									<SensitiveInput
-										placeholder={$i18n.t('API Key')}
-										bind:value={OPENAI_API_KEYS[idx]}
+						<hr class=" border-gray-50 dark:border-gray-850" />
+
+						<div class="">
+							<div class="flex justify-between items-center">
+								<div class="font-medium">{$i18n.t('Manage OpenAI API Connections')}</div>
+
+								<Tooltip content={$i18n.t(`Add Connection`)}>
+									<button
+										class="px-1"
+										on:click={() => {
+											showAddOpenAIConnectionModal = true;
+										}}
+										type="button"
+									>
+										<Plus />
+									</button>
+								</Tooltip>
+							</div>
+
+							<div class="flex flex-col gap-1.5 mt-1.5">
+								{#each OPENAI_API_BASE_URLS as url, idx}
+									<OpenAIConnection
+										pipeline={pipelineUrls[url] ? true : false}
+										bind:url
+										bind:key={OPENAI_API_KEYS[idx]}
+										bind:config={OPENAI_API_CONFIGS[url]}
+										onSubmit={() => {
+											updateOpenAIHandler();
+										}}
+										onDelete={() => {
+											OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS.filter(
+												(url, urlIdx) => idx !== urlIdx
+											);
+											OPENAI_API_KEYS = OPENAI_API_KEYS.filter((key, keyIdx) => idx !== keyIdx);
+										}}
 									/>
-									<div class="self-center flex items-center">
-										{#if idx === 0}
-											<button
-												class="px-1"
-												on:click={() => {
-													OPENAI_API_BASE_URLS = [...OPENAI_API_BASE_URLS, ''];
-													OPENAI_API_KEYS = [...OPENAI_API_KEYS, ''];
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														d="M8.75 3.75a.75.75 0 0 0-1.5 0v3.5h-3.5a.75.75 0 0 0 0 1.5h3.5v3.5a.75.75 0 0 0 1.5 0v-3.5h3.5a.75.75 0 0 0 0-1.5h-3.5v-3.5Z"
-													/>
-												</svg>
-											</button>
-										{:else}
-											<button
-												class="px-1"
-												on:click={() => {
-													OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS.filter(
-														(url, urlIdx) => idx !== urlIdx
-													);
-													OPENAI_API_KEYS = OPENAI_API_KEYS.filter((key, keyIdx) => idx !== keyIdx);
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path d="M3.75 7.25a.75.75 0 0 0 0 1.5h8.5a.75.75 0 0 0 0-1.5h-8.5Z" />
-												</svg>
-											</button>
-										{/if}
-									</div>
-
-									<div class="flex">
-										<Tooltip content="Verify connection" className="self-start mt-0.5">
-											<button
-												class="self-center p-2 bg-gray-200 hover:bg-gray-300 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
-												on:click={() => {
-													verifyOpenAIHandler(idx);
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 20 20"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														fill-rule="evenodd"
-														d="M15.312 11.424a5.5 5.5 0 01-9.201 2.466l-.312-.311h2.433a.75.75 0 000-1.5H3.989a.75.75 0 00-.75.75v4.242a.75.75 0 001.5 0v-2.43l.31.31a7 7 0 0011.712-3.138.75.75 0 00-1.449-.39zm1.23-3.723a.75.75 0 00.219-.53V2.929a.75.75 0 00-1.5 0V5.36l-.31-.31A7 7 0 003.239 8.188a.75.75 0 101.448.389A5.5 5.5 0 0113.89 6.11l.311.31h-2.432a.75.75 0 000 1.5h4.243a.75.75 0 00.53-.219z"
-														clip-rule="evenodd"
-													/>
-												</svg>
-											</button>
-										</Tooltip>
-									</div>
-								</div>
-								<div class=" mb-1 text-xs text-gray-400 dark:text-gray-500">
-									{$i18n.t('WebUI will make requests to')}
-									<span class=" text-gray-200">'{url}/models'</span>
-								</div>
-							{/each}
+								{/each}
+							</div>
 						</div>
 					{/if}
 				</div>
 			</div>
 
-			<hr class=" dark:border-gray-850" />
+			<hr class=" border-gray-50 dark:border-gray-850" />
 
-			<div class="pr-1.5 space-y-2">
-				<div class="flex justify-between items-center text-sm">
+			<div class="pr-1.5 my-2">
+				<div class="flex justify-between items-center text-sm mb-2">
 					<div class="  font-medium">{$i18n.t('Ollama API')}</div>
 
 					<div class="mt-1">
 						<Switch
 							bind:state={ENABLE_OLLAMA_API}
 							on:change={async () => {
-								updateOllamaConfig(localStorage.token, ENABLE_OLLAMA_API);
-
-								if (OLLAMA_BASE_URLS.length === 0) {
-									OLLAMA_BASE_URLS = [''];
-								}
+								updateOllamaHandler();
 							}}
 						/>
 					</div>
 				</div>
+
 				{#if ENABLE_OLLAMA_API}
-					<div class="flex w-full gap-1.5">
-						<div class="flex-1 flex flex-col gap-2">
-							{#each OLLAMA_BASE_URLS as url, idx}
-								<div class="flex gap-1.5">
-									<input
-										class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-										placeholder={$i18n.t('Enter URL (e.g. http://localhost:11434)')}
-										bind:value={url}
-									/>
+					<hr class=" border-gray-50 dark:border-gray-850 my-2" />
+
+					<div class="">
+						<div class="flex justify-between items-center">
+							<div class="font-medium">{$i18n.t('Manage Ollama API Connections')}</div>
+
+							<Tooltip content={$i18n.t(`Add Connection`)}>
+								<button
+									class="px-1"
+									on:click={() => {
+										showAddOllamaConnectionModal = true;
+									}}
+									type="button"
+								>
+									<Plus />
+								</button>
+							</Tooltip>
+						</div>
 
-									<div class="self-center flex items-center">
-										{#if idx === 0}
-											<button
-												class="px-1"
-												on:click={() => {
-													OLLAMA_BASE_URLS = [...OLLAMA_BASE_URLS, ''];
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														d="M8.75 3.75a.75.75 0 0 0-1.5 0v3.5h-3.5a.75.75 0 0 0 0 1.5h3.5v3.5a.75.75 0 0 0 1.5 0v-3.5h3.5a.75.75 0 0 0 0-1.5h-3.5v-3.5Z"
-													/>
-												</svg>
-											</button>
-										{:else}
-											<button
-												class="px-1"
-												on:click={() => {
-													OLLAMA_BASE_URLS = OLLAMA_BASE_URLS.filter(
-														(url, urlIdx) => idx !== urlIdx
-													);
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path d="M3.75 7.25a.75.75 0 0 0 0 1.5h8.5a.75.75 0 0 0 0-1.5h-8.5Z" />
-												</svg>
-											</button>
-										{/if}
-									</div>
-
-									<div class="flex">
-										<Tooltip content="Verify connection" className="self-start mt-0.5">
-											<button
-												class="self-center p-2 bg-gray-200 hover:bg-gray-300 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
-												on:click={() => {
-													verifyOllamaHandler(idx);
-												}}
-												type="button"
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 20 20"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														fill-rule="evenodd"
-														d="M15.312 11.424a5.5 5.5 0 01-9.201 2.466l-.312-.311h2.433a.75.75 0 000-1.5H3.989a.75.75 0 00-.75.75v4.242a.75.75 0 001.5 0v-2.43l.31.31a7 7 0 0011.712-3.138.75.75 0 00-1.449-.39zm1.23-3.723a.75.75 0 00.219-.53V2.929a.75.75 0 00-1.5 0V5.36l-.31-.31A7 7 0 003.239 8.188a.75.75 0 101.448.389A5.5 5.5 0 0113.89 6.11l.311.31h-2.432a.75.75 0 000 1.5h4.243a.75.75 0 00.53-.219z"
-														clip-rule="evenodd"
-													/>
-												</svg>
-											</button>
-										</Tooltip>
-									</div>
-								</div>
-							{/each}
+						<div class="flex w-full gap-1.5">
+							<div class="flex-1 flex flex-col gap-1.5 mt-1.5">
+								{#each OLLAMA_BASE_URLS as url, idx}
+									<OllamaConnection
+										bind:url
+										bind:config={OLLAMA_API_CONFIGS[url]}
+										{idx}
+										onSubmit={() => {
+											updateOllamaHandler();
+										}}
+										onDelete={() => {
+											OLLAMA_BASE_URLS = OLLAMA_BASE_URLS.filter((url, urlIdx) => idx !== urlIdx);
+										}}
+									/>
+								{/each}
+							</div>
 						</div>
-					</div>
 
-					<div class="mt-2 text-xs text-gray-400 dark:text-gray-500">
-						{$i18n.t('Trouble accessing Ollama?')}
-						<a
-							class=" text-gray-300 font-medium underline"
-							href="https://github.com/open-webui/open-webui#troubleshooting"
-							target="_blank"
-						>
-							{$i18n.t('Click here for help.')}
-						</a>
+						<div class="mt-1 text-xs text-gray-400 dark:text-gray-500">
+							{$i18n.t('Trouble accessing Ollama?')}
+							<a
+								class=" text-gray-300 font-medium underline"
+								href="https://github.com/open-webui/open-webui#troubleshooting"
+								target="_blank"
+							>
+								{$i18n.t('Click here for help.')}
+							</a>
+						</div>
 					</div>
 				{/if}
 			</div>

src/lib/components/admin/Settings/Connections/AddConnectionModal.svelte

@@ -0,0 +1,365 @@
+<script lang="ts">
+	import { toast } from 'svelte-sonner';
+	import { getContext, onMount } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import { models } from '$lib/stores';
+	import { verifyOpenAIConnection } from '$lib/apis/openai';
+	import { verifyOllamaConnection } from '$lib/apis/ollama';
+
+	import Modal from '$lib/components/common/Modal.svelte';
+	import Plus from '$lib/components/icons/Plus.svelte';
+	import Minus from '$lib/components/icons/Minus.svelte';
+	import PencilSolid from '$lib/components/icons/PencilSolid.svelte';
+	import SensitiveInput from '$lib/components/common/SensitiveInput.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import Switch from '$lib/components/common/Switch.svelte';
+
+	export let onSubmit: Function = () => {};
+	export let onDelete: Function = () => {};
+
+	export let show = false;
+	export let edit = false;
+	export let ollama = false;
+
+	export let connection = null;
+
+	let url = '';
+	let key = '';
+
+	let prefixId = '';
+	let enable = true;
+
+	let modelId = '';
+	let modelIds = [];
+
+	let loading = false;
+
+	const verifyOllamaHandler = async () => {
+		const res = await verifyOllamaConnection(localStorage.token, url, key).catch((error) => {
+			toast.error(error);
+		});
+
+		if (res) {
+			toast.success($i18n.t('Server connection verified'));
+		}
+	};
+
+	const verifyOpenAIHandler = async () => {
+		const res = await verifyOpenAIConnection(localStorage.token, url, key).catch((error) => {
+			toast.error(error);
+		});
+
+		if (res) {
+			toast.success($i18n.t('Server connection verified'));
+		}
+	};
+
+	const verifyHandler = () => {
+		if (ollama) {
+			verifyOllamaHandler();
+		} else {
+			verifyOpenAIHandler();
+		}
+	};
+
+	const addModelHandler = () => {
+		if (modelId) {
+			modelIds = [...modelIds, modelId];
+			modelId = '';
+		}
+	};
+
+	const submitHandler = async () => {
+		loading = true;
+
+		if (!ollama && (!url || !key)) {
+			loading = false;
+			toast.error('URL and Key are required');
+			return;
+		}
+
+		const connection = {
+			url,
+			key,
+			config: {
+				enable: enable,
+				prefix_id: prefixId,
+				model_ids: modelIds
+			}
+		};
+
+		await onSubmit(connection);
+
+		loading = false;
+		show = false;
+
+		url = '';
+		key = '';
+		prefixId = '';
+		modelIds = [];
+	};
+
+	const init = () => {
+		if (connection) {
+			url = connection.url;
+			key = connection.key;
+
+			enable = connection.config?.enable ?? true;
+			prefixId = connection.config?.prefix_id ?? '';
+			modelIds = connection.config?.model_ids ?? [];
+		}
+	};
+
+	$: if (show) {
+		init();
+	}
+
+	onMount(() => {
+		init();
+	});
+</script>
+
+<Modal size="sm" bind:show>
+	<div>
+		<div class=" flex justify-between dark:text-gray-100 px-5 pt-4 pb-2">
+			<div class=" text-lg font-medium self-center font-primary">
+				{#if edit}
+					{$i18n.t('Edit Connection')}
+				{:else}
+					{$i18n.t('Add Connection')}
+				{/if}
+			</div>
+			<button
+				class="self-center"
+				on:click={() => {
+					show = false;
+				}}
+			>
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 20 20"
+					fill="currentColor"
+					class="w-5 h-5"
+				>
+					<path
+						d="M6.28 5.22a.75.75 0 00-1.06 1.06L8.94 10l-3.72 3.72a.75.75 0 101.06 1.06L10 11.06l3.72 3.72a.75.75 0 101.06-1.06L11.06 10l3.72-3.72a.75.75 0 00-1.06-1.06L10 8.94 6.28 5.22z"
+					/>
+				</svg>
+			</button>
+		</div>
+
+		<div class="flex flex-col md:flex-row w-full px-4 pb-4 md:space-x-4 dark:text-gray-200">
+			<div class=" flex flex-col w-full sm:flex-row sm:justify-center sm:space-x-6">
+				<form
+					class="flex flex-col w-full"
+					on:submit={(e) => {
+						e.preventDefault();
+						submitHandler();
+					}}
+				>
+					<div class="px-1">
+						<div class="flex gap-2">
+							<div class="flex flex-col w-full">
+								<div class=" mb-0.5 text-xs text-gray-500">{$i18n.t('URL')}</div>
+
+								<div class="flex-1">
+									<input
+										class="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+										type="text"
+										bind:value={url}
+										placeholder={$i18n.t('API Base URL')}
+										autocomplete="off"
+										required
+									/>
+								</div>
+							</div>
+
+							<Tooltip content="Verify Connection" className="self-end -mb-1">
+								<button
+									class="self-center p-1 bg-transparent hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
+									on:click={() => {
+										verifyHandler();
+									}}
+									type="button"
+								>
+									<svg
+										xmlns="http://www.w3.org/2000/svg"
+										viewBox="0 0 20 20"
+										fill="currentColor"
+										class="w-4 h-4"
+									>
+										<path
+											fill-rule="evenodd"
+											d="M15.312 11.424a5.5 5.5 0 01-9.201 2.466l-.312-.311h2.433a.75.75 0 000-1.5H3.989a.75.75 0 00-.75.75v4.242a.75.75 0 001.5 0v-2.43l.31.31a7 7 0 0011.712-3.138.75.75 0 00-1.449-.39zm1.23-3.723a.75.75 0 00.219-.53V2.929a.75.75 0 00-1.5 0V5.36l-.31-.31A7 7 0 003.239 8.188a.75.75 0 101.448.389A5.5 5.5 0 0113.89 6.11l.311.31h-2.432a.75.75 0 000 1.5h4.243a.75.75 0 00.53-.219z"
+											clip-rule="evenodd"
+										/>
+									</svg>
+								</button>
+							</Tooltip>
+
+							<div class="flex flex-col flex-shrink-0 self-end">
+								<Tooltip content={enable ? $i18n.t('Enabled') : $i18n.t('Disabled')}>
+									<Switch bind:state={enable} />
+								</Tooltip>
+							</div>
+						</div>
+
+						<div class="flex gap-2 mt-2">
+							<div class="flex flex-col w-full">
+								<div class=" mb-0.5 text-xs text-gray-500">{$i18n.t('Key')}</div>
+
+								<div class="flex-1">
+									<SensitiveInput
+										className="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+										bind:value={key}
+										placeholder={$i18n.t('API Key')}
+										required={!ollama}
+									/>
+								</div>
+							</div>
+
+							<div class="flex flex-col w-full">
+								<div class=" mb-1 text-xs text-gray-500">{$i18n.t('Prefix ID')}</div>
+
+								<div class="flex-1">
+									<Tooltip
+										content={$i18n.t(
+											'Prefix ID is used to avoid conflicts with other connections by adding a prefix to the model IDs - leave empty to disable'
+										)}
+									>
+										<input
+											class="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+											type="text"
+											bind:value={prefixId}
+											placeholder={$i18n.t('Prefix ID')}
+											autocomplete="off"
+										/>
+									</Tooltip>
+								</div>
+							</div>
+						</div>
+
+						<hr class=" border-gray-100 dark:border-gray-700/10 my-2.5 w-full" />
+
+						<div class="flex flex-col w-full">
+							<div class="mb-1 flex justify-between">
+								<div class="text-xs text-gray-500">{$i18n.t('Model IDs')}</div>
+							</div>
+
+							{#if modelIds.length > 0}
+								<div class="flex flex-col">
+									{#each modelIds as modelId, modelIdx}
+										<div class=" flex gap-2 w-full justify-between items-center">
+											<div class=" text-sm flex-1 py-1 rounded-lg">
+												{modelId}
+											</div>
+											<div class="flex-shrink-0">
+												<button
+													type="button"
+													on:click={() => {
+														modelIds = modelIds.filter((_, idx) => idx !== modelIdx);
+													}}
+												>
+													<Minus strokeWidth="2" className="size-3.5" />
+												</button>
+											</div>
+										</div>
+									{/each}
+								</div>
+							{:else}
+								<div class="text-gray-500 text-xs text-center py-2 px-10">
+									{#if ollama}
+										{$i18n.t('Leave empty to include all models from "{{URL}}/api/tags" endpoint', {
+											URL: url
+										})}
+									{:else}
+										{$i18n.t('Leave empty to include all models from "{{URL}}/models" endpoint', {
+											URL: url
+										})}
+									{/if}
+								</div>
+							{/if}
+						</div>
+
+						<hr class=" border-gray-100 dark:border-gray-700/10 my-2.5 w-full" />
+
+						<div class="flex items-center">
+							<input
+								class="w-full py-1 text-sm rounded-lg bg-transparent {modelId
+									? ''
+									: 'text-gray-500'} placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+								bind:value={modelId}
+								placeholder={$i18n.t('Add a model ID')}
+							/>
+
+							<div>
+								<button
+									type="button"
+									on:click={() => {
+										addModelHandler();
+									}}
+								>
+									<Plus className="size-3.5" strokeWidth="2" />
+								</button>
+							</div>
+						</div>
+					</div>
+
+					<div class="flex justify-end pt-3 text-sm font-medium gap-1.5">
+						{#if edit}
+							<button
+								class="px-3.5 py-1.5 text-sm font-medium dark:bg-black dark:hover:bg-gray-900 dark:text-white bg-white text-black hover:bg-gray-100 transition rounded-full flex flex-row space-x-1 items-center"
+								type="button"
+								on:click={() => {
+									onDelete();
+									show = false;
+								}}
+							>
+								{$i18n.t('Delete')}
+							</button>
+						{/if}
+
+						<button
+							class="px-3.5 py-1.5 text-sm font-medium bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full flex flex-row space-x-1 items-center {loading
+								? ' cursor-not-allowed'
+								: ''}"
+							type="submit"
+							disabled={loading}
+						>
+							{$i18n.t('Save')}
+
+							{#if loading}
+								<div class="ml-2 self-center">
+									<svg
+										class=" w-4 h-4"
+										viewBox="0 0 24 24"
+										fill="currentColor"
+										xmlns="http://www.w3.org/2000/svg"
+										><style>
+											.spinner_ajPY {
+												transform-origin: center;
+												animation: spinner_AtaB 0.75s infinite linear;
+											}
+											@keyframes spinner_AtaB {
+												100% {
+													transform: rotate(360deg);
+												}
+											}
+										</style><path
+											d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
+											opacity=".25"
+										/><path
+											d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
+											class="spinner_ajPY"
+										/></svg
+									>
+								</div>
+							{/if}
+						</button>
+					</div>
+				</form>
+			</div>
+		</div>
+	</div>
+</Modal>

src/lib/components/admin/Settings/Connections/ManageOllamaModal.svelte

@@ -0,0 +1,1054 @@
+<script lang="ts">
+	import { toast } from 'svelte-sonner';
+	import { getContext, onMount } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import { WEBUI_NAME, models, MODEL_DOWNLOAD_POOL, user, config } from '$lib/stores';
+	import { splitStream } from '$lib/utils';
+
+	import {
+		createModel,
+		deleteModel,
+		downloadModel,
+		getOllamaUrls,
+		getOllamaVersion,
+		pullModel,
+		uploadModel,
+		getOllamaConfig,
+		getOllamaModels
+	} from '$lib/apis/ollama';
+	import { getModels } from '$lib/apis';
+
+	import Modal from '$lib/components/common/Modal.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import ModelDeleteConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
+	import Spinner from '$lib/components/common/Spinner.svelte';
+
+	export let show = false;
+
+	let modelUploadInputElement: HTMLInputElement;
+	let showModelDeleteConfirm = false;
+
+	let loading = true;
+
+	// Models
+	export let urlIdx: number | null = null;
+
+	let ollamaModels = [];
+
+	let updateModelId = null;
+	let updateProgress = null;
+	let showExperimentalOllama = false;
+
+	const MAX_PARALLEL_DOWNLOADS = 3;
+
+	let modelTransferring = false;
+	let modelTag = '';
+
+	let createModelLoading = false;
+	let createModelTag = '';
+	let createModelContent = '';
+	let createModelDigest = '';
+	let createModelPullProgress = null;
+
+	let digest = '';
+	let pullProgress = null;
+
+	let modelUploadMode = 'file';
+	let modelInputFile: File[] | null = null;
+	let modelFileUrl = '';
+	let modelFileContent = `TEMPLATE """{{ .System }}\nUSER: {{ .Prompt }}\nASSISTANT: """\nPARAMETER num_ctx 4096\nPARAMETER stop "</s>"\nPARAMETER stop "USER:"\nPARAMETER stop "ASSISTANT:"`;
+	let modelFileDigest = '';
+
+	let uploadProgress = null;
+	let uploadMessage = '';
+
+	let deleteModelTag = '';
+
+	const updateModelsHandler = async () => {
+		for (const model of ollamaModels) {
+			console.log(model);
+
+			updateModelId = model.id;
+			const [res, controller] = await pullModel(localStorage.token, model.id, urlIdx).catch(
+				(error) => {
+					toast.error(error);
+					return null;
+				}
+			);
+
+			if (res) {
+				const reader = res.body
+					.pipeThrough(new TextDecoderStream())
+					.pipeThrough(splitStream('\n'))
+					.getReader();
+
+				while (true) {
+					try {
+						const { value, done } = await reader.read();
+						if (done) break;
+
+						let lines = value.split('\n');
+
+						for (const line of lines) {
+							if (line !== '') {
+								let data = JSON.parse(line);
+
+								console.log(data);
+								if (data.error) {
+									throw data.error;
+								}
+								if (data.detail) {
+									throw data.detail;
+								}
+								if (data.status) {
+									if (data.digest) {
+										updateProgress = 0;
+										if (data.completed) {
+											updateProgress = Math.round((data.completed / data.total) * 1000) / 10;
+										} else {
+											updateProgress = 100;
+										}
+									} else {
+										toast.success(data.status);
+									}
+								}
+							}
+						}
+					} catch (error) {
+						console.log(error);
+					}
+				}
+			}
+		}
+
+		updateModelId = null;
+		updateProgress = null;
+	};
+
+	const pullModelHandler = async () => {
+		const sanitizedModelTag = modelTag.trim().replace(/^ollama\s+(run|pull)\s+/, '');
+		console.log($MODEL_DOWNLOAD_POOL);
+		if ($MODEL_DOWNLOAD_POOL[sanitizedModelTag]) {
+			toast.error(
+				$i18n.t(`Model '{{modelTag}}' is already in queue for downloading.`, {
+					modelTag: sanitizedModelTag
+				})
+			);
+			return;
+		}
+		if (Object.keys($MODEL_DOWNLOAD_POOL).length === MAX_PARALLEL_DOWNLOADS) {
+			toast.error(
+				$i18n.t('Maximum of 3 models can be downloaded simultaneously. Please try again later.')
+			);
+			return;
+		}
+
+		const [res, controller] = await pullModel(localStorage.token, sanitizedModelTag, urlIdx).catch(
+			(error) => {
+				toast.error(error);
+				return null;
+			}
+		);
+
+		if (res) {
+			const reader = res.body
+				.pipeThrough(new TextDecoderStream())
+				.pipeThrough(splitStream('\n'))
+				.getReader();
+
+			MODEL_DOWNLOAD_POOL.set({
+				...$MODEL_DOWNLOAD_POOL,
+				[sanitizedModelTag]: {
+					...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
+					abortController: controller,
+					reader,
+					done: false
+				}
+			});
+
+			while (true) {
+				try {
+					const { value, done } = await reader.read();
+					if (done) break;
+
+					let lines = value.split('\n');
+
+					for (const line of lines) {
+						if (line !== '') {
+							let data = JSON.parse(line);
+							console.log(data);
+							if (data.error) {
+								throw data.error;
+							}
+							if (data.detail) {
+								throw data.detail;
+							}
+
+							if (data.status) {
+								if (data.digest) {
+									let downloadProgress = 0;
+									if (data.completed) {
+										downloadProgress = Math.round((data.completed / data.total) * 1000) / 10;
+									} else {
+										downloadProgress = 100;
+									}
+
+									MODEL_DOWNLOAD_POOL.set({
+										...$MODEL_DOWNLOAD_POOL,
+										[sanitizedModelTag]: {
+											...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
+											pullProgress: downloadProgress,
+											digest: data.digest
+										}
+									});
+								} else {
+									toast.success(data.status);
+
+									MODEL_DOWNLOAD_POOL.set({
+										...$MODEL_DOWNLOAD_POOL,
+										[sanitizedModelTag]: {
+											...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
+											done: data.status === 'success'
+										}
+									});
+								}
+							}
+						}
+					}
+				} catch (error) {
+					console.log(error);
+					if (typeof error !== 'string') {
+						error = error.message;
+					}
+
+					toast.error(error);
+					// opts.callback({ success: false, error, modelName: opts.modelName });
+				}
+			}
+
+			console.log($MODEL_DOWNLOAD_POOL[sanitizedModelTag]);
+
+			if ($MODEL_DOWNLOAD_POOL[sanitizedModelTag].done) {
+				toast.success(
+					$i18n.t(`Model '{{modelName}}' has been successfully downloaded.`, {
+						modelName: sanitizedModelTag
+					})
+				);
+
+				models.set(await getModels(localStorage.token));
+			} else {
+				toast.error($i18n.t('Download canceled'));
+			}
+
+			delete $MODEL_DOWNLOAD_POOL[sanitizedModelTag];
+
+			MODEL_DOWNLOAD_POOL.set({
+				...$MODEL_DOWNLOAD_POOL
+			});
+		}
+
+		modelTag = '';
+		modelTransferring = false;
+	};
+
+	const uploadModelHandler = async () => {
+		modelTransferring = true;
+
+		let uploaded = false;
+		let fileResponse = null;
+		let name = '';
+
+		if (modelUploadMode === 'file') {
+			const file = modelInputFile ? modelInputFile[0] : null;
+
+			if (file) {
+				uploadMessage = 'Uploading...';
+
+				fileResponse = await uploadModel(localStorage.token, file, urlIdx).catch((error) => {
+					toast.error(error);
+					return null;
+				});
+			}
+		} else {
+			uploadProgress = 0;
+			fileResponse = await downloadModel(localStorage.token, modelFileUrl, urlIdx).catch(
+				(error) => {
+					toast.error(error);
+					return null;
+				}
+			);
+		}
+
+		if (fileResponse && fileResponse.ok) {
+			const reader = fileResponse.body
+				.pipeThrough(new TextDecoderStream())
+				.pipeThrough(splitStream('\n'))
+				.getReader();
+
+			while (true) {
+				const { value, done } = await reader.read();
+				if (done) break;
+
+				try {
+					let lines = value.split('\n');
+
+					for (const line of lines) {
+						if (line !== '') {
+							let data = JSON.parse(line.replace(/^data: /, ''));
+
+							if (data.progress) {
+								if (uploadMessage) {
+									uploadMessage = '';
+								}
+								uploadProgress = data.progress;
+							}
+
+							if (data.error) {
+								throw data.error;
+							}
+
+							if (data.done) {
+								modelFileDigest = data.blob;
+								name = data.name;
+								uploaded = true;
+							}
+						}
+					}
+				} catch (error) {
+					console.log(error);
+				}
+			}
+		} else {
+			const error = await fileResponse?.json();
+			toast.error(error?.detail ?? error);
+		}
+
+		if (uploaded) {
+			const res = await createModel(
+				localStorage.token,
+				`${name}:latest`,
+				`FROM @${modelFileDigest}\n${modelFileContent}`
+			);
+
+			if (res && res.ok) {
+				const reader = res.body
+					.pipeThrough(new TextDecoderStream())
+					.pipeThrough(splitStream('\n'))
+					.getReader();
+
+				while (true) {
+					const { value, done } = await reader.read();
+					if (done) break;
+
+					try {
+						let lines = value.split('\n');
+
+						for (const line of lines) {
+							if (line !== '') {
+								console.log(line);
+								let data = JSON.parse(line);
+								console.log(data);
+
+								if (data.error) {
+									throw data.error;
+								}
+								if (data.detail) {
+									throw data.detail;
+								}
+
+								if (data.status) {
+									if (
+										!data.digest &&
+										!data.status.includes('writing') &&
+										!data.status.includes('sha256')
+									) {
+										toast.success(data.status);
+									} else {
+										if (data.digest) {
+											digest = data.digest;
+
+											if (data.completed) {
+												pullProgress = Math.round((data.completed / data.total) * 1000) / 10;
+											} else {
+												pullProgress = 100;
+											}
+										}
+									}
+								}
+							}
+						}
+					} catch (error) {
+						console.log(error);
+						toast.error(error);
+					}
+				}
+			}
+		}
+
+		modelFileUrl = '';
+
+		if (modelUploadInputElement) {
+			modelUploadInputElement.value = '';
+		}
+		modelInputFile = null;
+		modelTransferring = false;
+		uploadProgress = null;
+
+		models.set(await getModels(localStorage.token));
+	};
+
+	const deleteModelHandler = async () => {
+		const res = await deleteModel(localStorage.token, deleteModelTag, urlIdx).catch((error) => {
+			toast.error(error);
+		});
+
+		if (res) {
+			toast.success($i18n.t(`Deleted {{deleteModelTag}}`, { deleteModelTag }));
+		}
+
+		deleteModelTag = '';
+		models.set(await getModels(localStorage.token));
+	};
+
+	const cancelModelPullHandler = async (model: string) => {
+		const { reader, abortController } = $MODEL_DOWNLOAD_POOL[model];
+		if (abortController) {
+			abortController.abort();
+		}
+		if (reader) {
+			await reader.cancel();
+			delete $MODEL_DOWNLOAD_POOL[model];
+			MODEL_DOWNLOAD_POOL.set({
+				...$MODEL_DOWNLOAD_POOL
+			});
+			await deleteModel(localStorage.token, model);
+			toast.success(`${model} download has been canceled`);
+		}
+	};
+
+	const createModelHandler = async () => {
+		createModelLoading = true;
+		const res = await createModel(
+			localStorage.token,
+			createModelTag,
+			createModelContent,
+			urlIdx
+		).catch((error) => {
+			toast.error(error);
+			return null;
+		});
+
+		if (res && res.ok) {
+			const reader = res.body
+				.pipeThrough(new TextDecoderStream())
+				.pipeThrough(splitStream('\n'))
+				.getReader();
+
+			while (true) {
+				const { value, done } = await reader.read();
+				if (done) break;
+
+				try {
+					let lines = value.split('\n');
+
+					for (const line of lines) {
+						if (line !== '') {
+							console.log(line);
+							let data = JSON.parse(line);
+							console.log(data);
+
+							if (data.error) {
+								throw data.error;
+							}
+							if (data.detail) {
+								throw data.detail;
+							}
+
+							if (data.status) {
+								if (
+									!data.digest &&
+									!data.status.includes('writing') &&
+									!data.status.includes('sha256')
+								) {
+									toast.success(data.status);
+								} else {
+									if (data.digest) {
+										createModelDigest = data.digest;
+
+										if (data.completed) {
+											createModelPullProgress =
+												Math.round((data.completed / data.total) * 1000) / 10;
+										} else {
+											createModelPullProgress = 100;
+										}
+									}
+								}
+							}
+						}
+					}
+				} catch (error) {
+					console.log(error);
+					toast.error(error);
+				}
+			}
+		}
+
+		models.set(await getModels(localStorage.token));
+
+		createModelLoading = false;
+
+		createModelTag = '';
+		createModelContent = '';
+		createModelDigest = '';
+		createModelPullProgress = null;
+	};
+
+	const init = async () => {
+		loading = true;
+		ollamaModels = await getOllamaModels(localStorage.token, urlIdx);
+
+		console.log(ollamaModels);
+		loading = false;
+	};
+
+	$: if (show) {
+		init();
+	}
+</script>
+
+<ModelDeleteConfirmDialog
+	bind:show={showModelDeleteConfirm}
+	on:confirm={() => {
+		deleteModelHandler();
+	}}
+/>
+
+<Modal size="sm" bind:show>
+	<div>
+		<div class=" flex justify-between dark:text-gray-100 px-5 pt-4 pb-2">
+			<div
+				class="flex w-full justify-between items-center text-lg font-medium self-center font-primary"
+			>
+				<div class=" flex-shrink-0">
+					{$i18n.t('Manage Ollama')}
+				</div>
+
+				<div>
+					<Tooltip content="Update All Models" placement="top">
+						<button
+							class="p-2.5 flex gap-2 items-center bg-transparent rounded-lg transition"
+							on:click={() => {
+								updateModelsHandler();
+							}}
+						>
+							<svg
+								xmlns="http://www.w3.org/2000/svg"
+								viewBox="0 0 16 16"
+								fill="currentColor"
+								class="w-4 h-4"
+							>
+								<path
+									d="M7 1a.75.75 0 0 1 .75.75V6h-1.5V1.75A.75.75 0 0 1 7 1ZM6.25 6v2.94L5.03 7.72a.75.75 0 0 0-1.06 1.06l2.5 2.5a.75.75 0 0 0 1.06 0l2.5-2.5a.75.75 0 1 0-1.06-1.06L7.75 8.94V6H10a2 2 0 0 1 2 2v3a2 2 0 0 1-2 2H4a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h2.25Z"
+								/>
+								<path
+									d="M4.268 14A2 2 0 0 0 6 15h6a2 2 0 0 0 2-2v-3a2 2 0 0 0-1-1.732V11a3 3 0 0 1-3 3H4.268Z"
+								/>
+							</svg>
+						</button>
+					</Tooltip>
+				</div>
+			</div>
+			<button
+				class="self-center"
+				on:click={() => {
+					show = false;
+				}}
+			>
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 20 20"
+					fill="currentColor"
+					class="w-5 h-5"
+				>
+					<path
+						d="M6.28 5.22a.75.75 0 00-1.06 1.06L8.94 10l-3.72 3.72a.75.75 0 101.06 1.06L10 11.06l3.72 3.72a.75.75 0 101.06-1.06L11.06 10l3.72-3.72a.75.75 0 00-1.06-1.06L10 8.94 6.28 5.22z"
+					/>
+				</svg>
+			</button>
+		</div>
+
+		<div class="flex flex-col md:flex-row w-full px-5 pb-4 md:space-x-4 dark:text-gray-200">
+			{#if !loading}
+				<div class=" flex flex-col w-full">
+					<div>
+						<div class="space-y-2">
+							{#if updateModelId}
+								<div class="text-xs">
+									Updating "{updateModelId}" {updateProgress ? `(${updateProgress}%)` : ''}
+								</div>
+							{/if}
+
+							<div>
+								<div class=" mb-2 text-sm font-medium">
+									{$i18n.t('Pull a model from Ollama.com')}
+								</div>
+								<div class="flex w-full">
+									<div class="flex-1 mr-2">
+										<input
+											class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+											placeholder={$i18n.t('Enter model tag (e.g. {{modelTag}})', {
+												modelTag: 'mistral:7b'
+											})}
+											bind:value={modelTag}
+										/>
+									</div>
+									<button
+										class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
+										on:click={() => {
+											pullModelHandler();
+										}}
+										disabled={modelTransferring}
+									>
+										{#if modelTransferring}
+											<div class="self-center">
+												<svg
+													class=" w-4 h-4"
+													viewBox="0 0 24 24"
+													fill="currentColor"
+													xmlns="http://www.w3.org/2000/svg"
+												>
+													<style>
+														.spinner_ajPY {
+															transform-origin: center;
+															animation: spinner_AtaB 0.75s infinite linear;
+														}
+
+														@keyframes spinner_AtaB {
+															100% {
+																transform: rotate(360deg);
+															}
+														}
+													</style>
+													<path
+														d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
+														opacity=".25"
+													/>
+													<path
+														d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
+														class="spinner_ajPY"
+													/>
+												</svg>
+											</div>
+										{:else}
+											<svg
+												xmlns="http://www.w3.org/2000/svg"
+												viewBox="0 0 16 16"
+												fill="currentColor"
+												class="w-4 h-4"
+											>
+												<path
+													d="M8.75 2.75a.75.75 0 0 0-1.5 0v5.69L5.03 6.22a.75.75 0 0 0-1.06 1.06l3.5 3.5a.75.75 0 0 0 1.06 0l3.5-3.5a.75.75 0 0 0-1.06-1.06L8.75 8.44V2.75Z"
+												/>
+												<path
+													d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
+												/>
+											</svg>
+										{/if}
+									</button>
+								</div>
+
+								<div class="mt-2 mb-1 text-xs text-gray-400 dark:text-gray-500">
+									{$i18n.t('To access the available model names for downloading,')}
+									<a
+										class=" text-gray-500 dark:text-gray-300 font-medium underline"
+										href="https://ollama.com/library"
+										target="_blank">{$i18n.t('click here.')}</a
+									>
+								</div>
+
+								{#if Object.keys($MODEL_DOWNLOAD_POOL).length > 0}
+									{#each Object.keys($MODEL_DOWNLOAD_POOL) as model}
+										{#if 'pullProgress' in $MODEL_DOWNLOAD_POOL[model]}
+											<div class="flex flex-col">
+												<div class="font-medium mb-1">{model}</div>
+												<div class="">
+													<div class="flex flex-row justify-between space-x-4 pr-2">
+														<div class=" flex-1">
+															<div
+																class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
+																style="width: {Math.max(
+																	15,
+																	$MODEL_DOWNLOAD_POOL[model].pullProgress ?? 0
+																)}%"
+															>
+																{$MODEL_DOWNLOAD_POOL[model].pullProgress ?? 0}%
+															</div>
+														</div>
+
+														<Tooltip content={$i18n.t('Cancel')}>
+															<button
+																class="text-gray-800 dark:text-gray-100"
+																on:click={() => {
+																	cancelModelPullHandler(model);
+																}}
+															>
+																<svg
+																	class="w-4 h-4 text-gray-800 dark:text-white"
+																	aria-hidden="true"
+																	xmlns="http://www.w3.org/2000/svg"
+																	width="24"
+																	height="24"
+																	fill="currentColor"
+																	viewBox="0 0 24 24"
+																>
+																	<path
+																		stroke="currentColor"
+																		stroke-linecap="round"
+																		stroke-linejoin="round"
+																		stroke-width="2"
+																		d="M6 18 17.94 6M18 18 6.06 6"
+																	/>
+																</svg>
+															</button>
+														</Tooltip>
+													</div>
+													{#if 'digest' in $MODEL_DOWNLOAD_POOL[model]}
+														<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
+															{$MODEL_DOWNLOAD_POOL[model].digest}
+														</div>
+													{/if}
+												</div>
+											</div>
+										{/if}
+									{/each}
+								{/if}
+							</div>
+
+							<div>
+								<div class=" mb-2 text-sm font-medium">{$i18n.t('Delete a model')}</div>
+								<div class="flex w-full">
+									<div class="flex-1 mr-2">
+										<select
+											class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+											bind:value={deleteModelTag}
+											placeholder={$i18n.t('Select a model')}
+										>
+											{#if !deleteModelTag}
+												<option value="" disabled selected>{$i18n.t('Select a model')}</option>
+											{/if}
+											{#each ollamaModels as model}
+												<option value={model.id} class="bg-gray-50 dark:bg-gray-700"
+													>{model.name +
+														' (' +
+														(model.size / 1024 ** 3).toFixed(1) +
+														' GB)'}</option
+												>
+											{/each}
+										</select>
+									</div>
+									<button
+										class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
+										on:click={() => {
+											showModelDeleteConfirm = true;
+										}}
+									>
+										<svg
+											xmlns="http://www.w3.org/2000/svg"
+											viewBox="0 0 16 16"
+											fill="currentColor"
+											class="w-4 h-4"
+										>
+											<path
+												fill-rule="evenodd"
+												d="M5 3.25V4H2.75a.75.75 0 0 0 0 1.5h.3l.815 8.15A1.5 1.5 0 0 0 5.357 15h5.285a1.5 1.5 0 0 0 1.493-1.35l.815-8.15h.3a.75.75 0 0 0 0-1.5H11v-.75A2.25 2.25 0 0 0 8.75 1h-1.5A2.25 2.25 0 0 0 5 3.25Zm2.25-.75a.75.75 0 0 0-.75.75V4h3v-.75a.75.75 0 0 0-.75-.75h-1.5ZM6.05 6a.75.75 0 0 1 .787.713l.275 5.5a.75.75 0 0 1-1.498.075l-.275-5.5A.75.75 0 0 1 6.05 6Zm3.9 0a.75.75 0 0 1 .712.787l-.275 5.5a.75.75 0 0 1-1.498-.075l.275-5.5a.75.75 0 0 1 .786-.711Z"
+												clip-rule="evenodd"
+											/>
+										</svg>
+									</button>
+								</div>
+							</div>
+
+							<div>
+								<div class=" mb-2 text-sm font-medium">{$i18n.t('Create a model')}</div>
+								<div class="flex w-full">
+									<div class="flex-1 mr-2 flex flex-col gap-2">
+										<input
+											class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+											placeholder={$i18n.t('Enter model tag (e.g. {{modelTag}})', {
+												modelTag: 'my-modelfile'
+											})}
+											bind:value={createModelTag}
+											disabled={createModelLoading}
+										/>
+
+										<textarea
+											bind:value={createModelContent}
+											class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-100 dark:bg-gray-850 outline-none resize-none scrollbar-hidden"
+											rows="6"
+											placeholder={`TEMPLATE """{{ .System }}\nUSER: {{ .Prompt }}\nASSISTANT: """\nPARAMETER num_ctx 4096\nPARAMETER stop "</s>"\nPARAMETER stop "USER:"\nPARAMETER stop "ASSISTANT:"`}
+											disabled={createModelLoading}
+										/>
+									</div>
+
+									<div class="flex self-start">
+										<button
+											class="px-2.5 py-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition disabled:cursor-not-allowed"
+											on:click={() => {
+												createModelHandler();
+											}}
+											disabled={createModelLoading}
+										>
+											<svg
+												xmlns="http://www.w3.org/2000/svg"
+												viewBox="0 0 16 16"
+												fill="currentColor"
+												class="size-4"
+											>
+												<path
+													d="M7.25 10.25a.75.75 0 0 0 1.5 0V4.56l2.22 2.22a.75.75 0 1 0 1.06-1.06l-3.5-3.5a.75.75 0 0 0-1.06 0l-3.5 3.5a.75.75 0 0 0 1.06 1.06l2.22-2.22v5.69Z"
+												/>
+												<path
+													d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
+												/>
+											</svg>
+										</button>
+									</div>
+								</div>
+
+								{#if createModelDigest !== ''}
+									<div class="flex flex-col mt-1">
+										<div class="font-medium mb-1">{createModelTag}</div>
+										<div class="">
+											<div class="flex flex-row justify-between space-x-4 pr-2">
+												<div class=" flex-1">
+													<div
+														class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
+														style="width: {Math.max(15, createModelPullProgress ?? 0)}%"
+													>
+														{createModelPullProgress ?? 0}%
+													</div>
+												</div>
+											</div>
+											{#if createModelDigest}
+												<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
+													{createModelDigest}
+												</div>
+											{/if}
+										</div>
+									</div>
+								{/if}
+							</div>
+
+							<div class="pt-1">
+								<div class="flex justify-between items-center text-xs">
+									<div class=" text-sm font-medium">{$i18n.t('Experimental')}</div>
+									<button
+										class=" text-xs font-medium text-gray-500"
+										type="button"
+										on:click={() => {
+											showExperimentalOllama = !showExperimentalOllama;
+										}}>{showExperimentalOllama ? $i18n.t('Hide') : $i18n.t('Show')}</button
+									>
+								</div>
+							</div>
+
+							{#if showExperimentalOllama}
+								<form
+									on:submit|preventDefault={() => {
+										uploadModelHandler();
+									}}
+								>
+									<div class=" mb-2 flex w-full justify-between">
+										<div class="  text-sm font-medium">{$i18n.t('Upload a GGUF model')}</div>
+
+										<button
+											class="p-1 px-3 text-xs flex rounded transition"
+											on:click={() => {
+												if (modelUploadMode === 'file') {
+													modelUploadMode = 'url';
+												} else {
+													modelUploadMode = 'file';
+												}
+											}}
+											type="button"
+										>
+											{#if modelUploadMode === 'file'}
+												<span class="ml-2 self-center">{$i18n.t('File Mode')}</span>
+											{:else}
+												<span class="ml-2 self-center">{$i18n.t('URL Mode')}</span>
+											{/if}
+										</button>
+									</div>
+
+									<div class="flex w-full mb-1.5">
+										<div class="flex flex-col w-full">
+											{#if modelUploadMode === 'file'}
+												<div
+													class="flex-1 {modelInputFile && modelInputFile.length > 0 ? 'mr-2' : ''}"
+												>
+													<input
+														id="model-upload-input"
+														bind:this={modelUploadInputElement}
+														type="file"
+														bind:files={modelInputFile}
+														on:change={() => {
+															console.log(modelInputFile);
+														}}
+														accept=".gguf,.safetensors"
+														required
+														hidden
+													/>
+
+													<button
+														type="button"
+														class="w-full rounded-lg text-left py-2 px-4 bg-gray-50 dark:text-gray-300 dark:bg-gray-850"
+														on:click={() => {
+															modelUploadInputElement.click();
+														}}
+													>
+														{#if modelInputFile && modelInputFile.length > 0}
+															{modelInputFile[0].name}
+														{:else}
+															{$i18n.t('Click here to select')}
+														{/if}
+													</button>
+												</div>
+											{:else}
+												<div class="flex-1 {modelFileUrl !== '' ? 'mr-2' : ''}">
+													<input
+														class="w-full rounded-lg text-left py-2 px-4 bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none {modelFileUrl !==
+														''
+															? 'mr-2'
+															: ''}"
+														type="url"
+														required
+														bind:value={modelFileUrl}
+														placeholder={$i18n.t('Type Hugging Face Resolve (Download) URL')}
+													/>
+												</div>
+											{/if}
+										</div>
+
+										{#if (modelUploadMode === 'file' && modelInputFile && modelInputFile.length > 0) || (modelUploadMode === 'url' && modelFileUrl !== '')}
+											<button
+												class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg disabled:cursor-not-allowed transition"
+												type="submit"
+												disabled={modelTransferring}
+											>
+												{#if modelTransferring}
+													<div class="self-center">
+														<svg
+															class=" w-4 h-4"
+															viewBox="0 0 24 24"
+															fill="currentColor"
+															xmlns="http://www.w3.org/2000/svg"
+														>
+															<style>
+																.spinner_ajPY {
+																	transform-origin: center;
+																	animation: spinner_AtaB 0.75s infinite linear;
+																}
+
+																@keyframes spinner_AtaB {
+																	100% {
+																		transform: rotate(360deg);
+																	}
+																}
+															</style>
+															<path
+																d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
+																opacity=".25"
+															/>
+															<path
+																d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
+																class="spinner_ajPY"
+															/>
+														</svg>
+													</div>
+												{:else}
+													<svg
+														xmlns="http://www.w3.org/2000/svg"
+														viewBox="0 0 16 16"
+														fill="currentColor"
+														class="w-4 h-4"
+													>
+														<path
+															d="M7.25 10.25a.75.75 0 0 0 1.5 0V4.56l2.22 2.22a.75.75 0 1 0 1.06-1.06l-3.5-3.5a.75.75 0 0 0-1.06 0l-3.5 3.5a.75.75 0 0 0 1.06 1.06l2.22-2.22v5.69Z"
+														/>
+														<path
+															d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
+														/>
+													</svg>
+												{/if}
+											</button>
+										{/if}
+									</div>
+
+									{#if (modelUploadMode === 'file' && modelInputFile && modelInputFile.length > 0) || (modelUploadMode === 'url' && modelFileUrl !== '')}
+										<div>
+											<div>
+												<div class=" my-2.5 text-sm font-medium">
+													{$i18n.t('Modelfile Content')}
+												</div>
+												<textarea
+													bind:value={modelFileContent}
+													class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-100 dark:bg-gray-850 outline-none resize-none"
+													rows="6"
+												/>
+											</div>
+										</div>
+									{/if}
+									<div class=" mt-1 text-xs text-gray-400 dark:text-gray-500">
+										{$i18n.t('To access the GGUF models available for downloading,')}
+										<a
+											class=" text-gray-500 dark:text-gray-300 font-medium underline"
+											href="https://huggingface.co/models?search=gguf"
+											target="_blank">{$i18n.t('click here.')}</a
+										>
+									</div>
+
+									{#if uploadMessage}
+										<div class="mt-2">
+											<div class=" mb-2 text-xs">{$i18n.t('Upload Progress')}</div>
+
+											<div class="w-full rounded-full dark:bg-gray-800">
+												<div
+													class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
+													style="width: 100%"
+												>
+													{uploadMessage}
+												</div>
+											</div>
+											<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
+												{modelFileDigest}
+											</div>
+										</div>
+									{:else if uploadProgress !== null}
+										<div class="mt-2">
+											<div class=" mb-2 text-xs">{$i18n.t('Upload Progress')}</div>
+
+											<div class="w-full rounded-full dark:bg-gray-800">
+												<div
+													class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
+													style="width: {Math.max(15, uploadProgress ?? 0)}%"
+												>
+													{uploadProgress ?? 0}%
+												</div>
+											</div>
+											<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
+												{modelFileDigest}
+											</div>
+										</div>
+									{/if}
+								</form>
+							{/if}
+						</div>
+					</div>
+				</div>
+			{:else}
+				<Spinner />
+			{/if}
+		</div>
+	</div>
+</Modal>

src/lib/components/admin/Settings/Connections/OllamaConnection.svelte

@@ -0,0 +1,89 @@
+<script lang="ts">
+	import { getContext, tick } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import SensitiveInput from '$lib/components/common/SensitiveInput.svelte';
+	import AddConnectionModal from './AddConnectionModal.svelte';
+
+	import Cog6 from '$lib/components/icons/Cog6.svelte';
+	import Wrench from '$lib/components/icons/Wrench.svelte';
+	import ManageOllamaModal from './ManageOllamaModal.svelte';
+
+	export let onDelete = () => {};
+	export let onSubmit = () => {};
+
+	export let url = '';
+	export let idx = 0;
+	export let config = {};
+
+	let showManageModal = false;
+	let showConfigModal = false;
+</script>
+
+<AddConnectionModal
+	ollama
+	edit
+	bind:show={showConfigModal}
+	connection={{
+		url,
+		key: config?.key ?? '',
+		config: config
+	}}
+	{onDelete}
+	onSubmit={(connection) => {
+		url = connection.url;
+		config = { ...connection.config, key: connection.key };
+		onSubmit(connection);
+	}}
+/>
+
+<ManageOllamaModal bind:show={showManageModal} urlIdx={idx} />
+
+<div class="flex gap-1.5">
+	<Tooltip
+		className="w-full relative"
+		content={$i18n.t(`WebUI will make requests to "{{url}}/api/chat"`, {
+			url
+		})}
+		placement="top-start"
+	>
+		{#if !(config?.enable ?? true)}
+			<div
+				class="absolute top-0 bottom-0 left-0 right-0 opacity-60 bg-white dark:bg-gray-900 z-10"
+			></div>
+		{/if}
+
+		<input
+			class="w-full text-sm bg-transparent outline-none"
+			placeholder={$i18n.t('Enter URL (e.g. http://localhost:11434)')}
+			bind:value={url}
+		/>
+	</Tooltip>
+
+	<div class="flex gap-1">
+		<Tooltip content={$i18n.t('Manage')} className="self-start">
+			<button
+				class="self-center p-1 bg-transparent hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
+				on:click={() => {
+					showManageModal = true;
+				}}
+				type="button"
+			>
+				<Wrench />
+			</button>
+		</Tooltip>
+
+		<Tooltip content={$i18n.t('Configure')} className="self-start">
+			<button
+				class="self-center p-1 bg-transparent hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
+				on:click={() => {
+					showConfigModal = true;
+				}}
+				type="button"
+			>
+				<Cog6 />
+			</button>
+		</Tooltip>
+	</div>
+</div>

src/lib/components/admin/Settings/Connections/OpenAIConnection.svelte

@@ -0,0 +1,107 @@
+<script lang="ts">
+	import { getContext, tick } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import SensitiveInput from '$lib/components/common/SensitiveInput.svelte';
+	import Cog6 from '$lib/components/icons/Cog6.svelte';
+	import AddConnectionModal from './AddConnectionModal.svelte';
+	import { connect } from 'socket.io-client';
+
+	export let onDelete = () => {};
+	export let onSubmit = () => {};
+
+	export let pipeline = false;
+
+	export let url = '';
+	export let key = '';
+	export let config = {};
+
+	let showConfigModal = false;
+</script>
+
+<AddConnectionModal
+	edit
+	bind:show={showConfigModal}
+	connection={{
+		url,
+		key,
+		config
+	}}
+	{onDelete}
+	onSubmit={(connection) => {
+		url = connection.url;
+		key = connection.key;
+		config = connection.config;
+		onSubmit(connection);
+	}}
+/>
+
+<div class="flex w-full gap-2 items-center">
+	<Tooltip
+		className="w-full relative"
+		content={$i18n.t(`WebUI will make requests to "{{url}}/chat/completions"`, {
+			url
+		})}
+		placement="top-start"
+	>
+		{#if !(config?.enable ?? true)}
+			<div
+				class="absolute top-0 bottom-0 left-0 right-0 opacity-60 bg-white dark:bg-gray-900 z-10"
+			></div>
+		{/if}
+		<div class="flex w-full">
+			<div class="flex-1 relative">
+				<input
+					class=" outline-none w-full bg-transparent {pipeline ? 'pr-8' : ''}"
+					placeholder={$i18n.t('API Base URL')}
+					bind:value={url}
+					autocomplete="off"
+				/>
+
+				{#if pipeline}
+					<div class=" absolute top-2.5 right-2.5">
+						<Tooltip content="Pipelines">
+							<svg
+								xmlns="http://www.w3.org/2000/svg"
+								viewBox="0 0 24 24"
+								fill="currentColor"
+								class="size-4"
+							>
+								<path
+									d="M11.644 1.59a.75.75 0 0 1 .712 0l9.75 5.25a.75.75 0 0 1 0 1.32l-9.75 5.25a.75.75 0 0 1-.712 0l-9.75-5.25a.75.75 0 0 1 0-1.32l9.75-5.25Z"
+								/>
+								<path
+									d="m3.265 10.602 7.668 4.129a2.25 2.25 0 0 0 2.134 0l7.668-4.13 1.37.739a.75.75 0 0 1 0 1.32l-9.75 5.25a.75.75 0 0 1-.71 0l-9.75-5.25a.75.75 0 0 1 0-1.32l1.37-.738Z"
+								/>
+								<path
+									d="m10.933 19.231-7.668-4.13-1.37.739a.75.75 0 0 0 0 1.32l9.75 5.25c.221.12.489.12.71 0l9.75-5.25a.75.75 0 0 0 0-1.32l-1.37-.738-7.668 4.13a2.25 2.25 0 0 1-2.134-.001Z"
+								/>
+							</svg>
+						</Tooltip>
+					</div>
+				{/if}
+			</div>
+
+			<SensitiveInput
+				inputClassName=" outline-none bg-transparent w-full"
+				placeholder={$i18n.t('API Key')}
+				bind:value={key}
+			/>
+		</div>
+	</Tooltip>
+
+	<div class="flex gap-1">
+		<Tooltip content={$i18n.t('Configure')} className="self-start">
+			<button
+				class="self-center p-1 bg-transparent hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 rounded-lg transition"
+				on:click={() => {
+					showConfigModal = true;
+				}}
+				type="button"
+			>
+				<Cog6 />
+			</button>
+		</Tooltip>
+	</div>
+</div>

src/lib/components/admin/Settings/Database.svelte

@@ -181,37 +181,6 @@
 					</div>
 				</button>
 			{/if}
-
-			<hr class=" dark:border-gray-850 my-1" />
-
-			<button
-				type="button"
-				class=" flex rounded-md py-2 px-3 w-full hover:bg-gray-200 dark:hover:bg-gray-800 transition"
-				on:click={() => {
-					downloadLiteLLMConfig(localStorage.token).catch((error) => {
-						toast.error(error);
-					});
-				}}
-			>
-				<div class=" self-center mr-3">
-					<svg
-						xmlns="http://www.w3.org/2000/svg"
-						viewBox="0 0 16 16"
-						fill="currentColor"
-						class="w-4 h-4"
-					>
-						<path d="M2 3a1 1 0 0 1 1-1h10a1 1 0 0 1 1 1v1a1 1 0 0 1-1 1H3a1 1 0 0 1-1-1V3Z" />
-						<path
-							fill-rule="evenodd"
-							d="M13 6H3v6a2 2 0 0 0 2 2h6a2 2 0 0 0 2-2V6ZM8.75 7.75a.75.75 0 0 0-1.5 0v2.69L6.03 9.22a.75.75 0 0 0-1.06 1.06l2.5 2.5a.75.75 0 0 0 1.06 0l2.5-2.5a.75.75 0 1 0-1.06-1.06l-1.22 1.22V7.75Z"
-							clip-rule="evenodd"
-						/>
-					</svg>
-				</div>
-				<div class=" self-center text-sm font-medium">
-					{$i18n.t('Export LiteLLM config.yaml')}
-				</div>
-			</button>
 		</div>
 	</div>
 

src/lib/components/admin/Settings/Documents.svelte

@@ -19,7 +19,7 @@
 	} from '$lib/apis/retrieval';
 
 	import { knowledge, models } from '$lib/stores';
-	import { getKnowledgeItems } from '$lib/apis/knowledge';
+	import { getKnowledgeBases } from '$lib/apis/knowledge';
 	import { uploadDir, deleteAllFiles, deleteFileById } from '$lib/apis/files';
 
 	import ResetUploadDirConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
@@ -56,8 +56,11 @@
 	let chunkOverlap = 0;
 	let pdfExtractImages = true;
 
-	let OpenAIKey = '';
 	let OpenAIUrl = '';
+	let OpenAIKey = '';
+
+	let OllamaUrl = '';
+	let OllamaKey = '';
 
 	let querySettings = {
 		template: '',
@@ -104,19 +107,15 @@
 		const res = await updateEmbeddingConfig(localStorage.token, {
 			embedding_engine: embeddingEngine,
 			embedding_model: embeddingModel,
-			...(embeddingEngine === 'openai' || embeddingEngine === 'ollama'
-				? {
-						embedding_batch_size: embeddingBatchSize
-					}
-				: {}),
-			...(embeddingEngine === 'openai'
-				? {
-						openai_config: {
-							key: OpenAIKey,
-							url: OpenAIUrl
-						}
-					}
-				: {})
+			embedding_batch_size: embeddingBatchSize,
+			ollama_config: {
+				key: OllamaKey,
+				url: OllamaUrl
+			},
+			openai_config: {
+				key: OpenAIKey,
+				url: OpenAIUrl
+			}
 		}).catch(async (error) => {
 			toast.error(error);
 			await setEmbeddingConfig();
@@ -206,6 +205,9 @@
 
 			OpenAIKey = embeddingConfig.openai_config.key;
 			OpenAIUrl = embeddingConfig.openai_config.url;
+
+			OllamaKey = embeddingConfig.ollama_config.key;
+			OllamaUrl = embeddingConfig.ollama_config.url;
 		}
 	};
 
@@ -310,9 +312,9 @@
 			</div>
 
 			{#if embeddingEngine === 'openai'}
-				<div class="my-0.5 flex gap-2">
+				<div class="my-0.5 flex gap-2 pr-2">
 					<input
-						class="flex-1 w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+						class="flex-1 w-full rounded-lg text-sm bg-transparent outline-none"
 						placeholder={$i18n.t('API Base URL')}
 						bind:value={OpenAIUrl}
 						required
@@ -320,7 +322,23 @@
 
 					<SensitiveInput placeholder={$i18n.t('API Key')} bind:value={OpenAIKey} />
 				</div>
+			{:else if embeddingEngine === 'ollama'}
+				<div class="my-0.5 flex gap-2 pr-2">
+					<input
+						class="flex-1 w-full rounded-lg text-sm bg-transparent outline-none"
+						placeholder={$i18n.t('API Base URL')}
+						bind:value={OllamaUrl}
+						required
+					/>
+
+					<SensitiveInput
+						placeholder={$i18n.t('API Key')}
+						bind:value={OllamaKey}
+						required={false}
+					/>
+				</div>
 			{/if}
+
 			{#if embeddingEngine === 'ollama' || embeddingEngine === 'openai'}
 				<div class="flex mt-0.5 space-x-2">
 					<div class=" self-center text-xs font-medium">{$i18n.t('Embedding Batch Size')}</div>
@@ -376,19 +394,12 @@
 			{#if embeddingEngine === 'ollama'}
 				<div class="flex w-full">
 					<div class="flex-1 mr-2">
-						<select
+						<input
 							class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
 							bind:value={embeddingModel}
-							placeholder={$i18n.t('Select a model')}
+							placeholder={$i18n.t('Set embedding model')}
 							required
-						>
-							{#if !embeddingModel}
-								<option value="" disabled selected>{$i18n.t('Select a model')}</option>
-							{/if}
-							{#each $models.filter((m) => m.id && m.ollama && !(m?.preset ?? false)) as model}
-								<option value={model.id} class="bg-gray-50 dark:bg-gray-700">{model.name}</option>
-							{/each}
-						</select>
+						/>
 					</div>
 				</div>
 			{:else}

src/lib/components/admin/Settings/Evaluations.svelte

@@ -11,7 +11,7 @@
 	import Tooltip from '$lib/components/common/Tooltip.svelte';
 	import Plus from '$lib/components/icons/Plus.svelte';
 	import Model from './Evaluations/Model.svelte';
-	import ModelModal from './Evaluations/ModelModal.svelte';
+	import ArenaModelModal from './Evaluations/ArenaModelModal.svelte';
 	import { getConfig, updateConfig } from '$lib/apis/evaluations';
 
 	const i18n = getContext('i18n');
@@ -65,7 +65,7 @@
 	});
 </script>
 
-<ModelModal
+<ArenaModelModal
 	bind:show={showAddModel}
 	on:submit={async (e) => {
 		addModelHandler(e.detail);

src/lib/components/admin/Settings/Evaluations/ModelModal.svelte → src/lib/components/admin/Settings/Evaluations/ArenaModelModal.svelte

@@ -9,6 +9,7 @@
 	import Minus from '$lib/components/icons/Minus.svelte';
 	import PencilSolid from '$lib/components/icons/PencilSolid.svelte';
 	import { toast } from 'svelte-sonner';
+	import AccessControl from '$lib/components/workspace/common/AccessControl.svelte';
 
 	export let show = false;
 	export let edit = false;
@@ -39,6 +40,8 @@
 	let modelIds = [];
 	let filterMode = 'include';
 
+	let accessControl = {};
+
 	let imageInputElement;
 	let loading = false;
 
@@ -74,7 +77,8 @@
 				profile_image_url: profileImageUrl,
 				description: description || null,
 				model_ids: modelIds.length > 0 ? modelIds : null,
-				filter_mode: modelIds.length > 0 ? (filterMode ? filterMode : null) : null
+				filter_mode: modelIds.length > 0 ? (filterMode ? filterMode : null) : null,
+				access_control: accessControl
 			}
 		};
 
@@ -98,6 +102,7 @@
 			description = model.meta.description;
 			modelIds = model.meta.model_ids || [];
 			filterMode = model.meta?.filter_mode ?? 'include';
+			accessControl = 'access_control' in model.meta ? model.meta.access_control : {};
 		}
 	};
 
@@ -283,6 +288,14 @@
 
 						<hr class=" border-gray-100 dark:border-gray-700/10 my-2.5 w-full" />
 
+						<div class="my-2 -mx-2">
+							<div class="px-3 py-2 bg-gray-50 dark:bg-gray-950 rounded-lg">
+								<AccessControl bind:accessControl />
+							</div>
+						</div>
+
+						<hr class=" border-gray-100 dark:border-gray-700/10 my-2.5 w-full" />
+
 						<div class="flex flex-col w-full">
 							<div class="mb-1 flex justify-between">
 								<div class="text-xs text-gray-500">{$i18n.t('Models')}</div>

src/lib/components/admin/Settings/Evaluations/Model.svelte

@@ -4,13 +4,13 @@
 	const i18n = getContext('i18n');
 
 	import Cog6 from '$lib/components/icons/Cog6.svelte';
-	import ModelModal from './ModelModal.svelte';
+	import ArenaModelModal from './ArenaModelModal.svelte';
 	export let model;
 
 	let showModel = false;
 </script>
 
-<ModelModal
+<ArenaModelModal
 	bind:show={showModel}
 	edit={true}
 	{model}

src/lib/components/admin/Settings/General.svelte

@@ -1,7 +1,16 @@
 <script lang="ts">
 	import { getBackendConfig, getWebhookUrl, updateWebhookUrl } from '$lib/apis';
-	import { getAdminConfig, updateAdminConfig } from '$lib/apis/auths';
+	import {
+		getAdminConfig,
+		getLdapConfig,
+		getLdapServer,
+		updateAdminConfig,
+		updateLdapConfig,
+		updateLdapServer
+	} from '$lib/apis/auths';
+	import SensitiveInput from '$lib/components/common/SensitiveInput.svelte';
 	import Switch from '$lib/components/common/Switch.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
 	import { config } from '$lib/stores';
 	import { onMount, getContext } from 'svelte';
 	import { toast } from 'svelte-sonner';
@@ -13,9 +22,37 @@
 	let adminConfig = null;
 	let webhookUrl = '';
 
+	// LDAP
+	let ENABLE_LDAP = false;
+	let LDAP_SERVER = {
+		label: '',
+		host: '',
+		port: '',
+		attribute_for_username: 'uid',
+		app_dn: '',
+		app_dn_password: '',
+		search_base: '',
+		search_filters: '',
+		use_tls: false,
+		certificate_path: '',
+		ciphers: ''
+	};
+
+	const updateLdapServerHandler = async () => {
+		if (!ENABLE_LDAP) return;
+		const res = await updateLdapServer(localStorage.token, LDAP_SERVER).catch((error) => {
+			toast.error(error);
+			return null;
+		});
+		if (res) {
+			toast.success($i18n.t('LDAP server updated'));
+		}
+	};
+
 	const updateHandler = async () => {
 		webhookUrl = await updateWebhookUrl(localStorage.token, webhookUrl);
 		const res = await updateAdminConfig(localStorage.token, adminConfig);
+		await updateLdapServerHandler();
 
 		if (res) {
 			saveHandler();
@@ -32,8 +69,14 @@
 
 			(async () => {
 				webhookUrl = await getWebhookUrl(localStorage.token);
+			})(),
+			(async () => {
+				LDAP_SERVER = await getLdapServer(localStorage.token);
 			})()
 		]);
+
+		const ldapConfig = await getLdapConfig(localStorage.token);
+		ENABLE_LDAP = ldapConfig.ENABLE_LDAP;
 	});
 </script>
 
@@ -69,7 +112,13 @@
 					</div>
 				</div>
 
-				<hr class=" dark:border-gray-850 my-2" />
+				<div class="  flex w-full justify-between pr-2">
+					<div class=" self-center text-xs font-medium">{$i18n.t('Enable API Key Auth')}</div>
+
+					<Switch bind:state={adminConfig.ENABLE_API_KEY} />
+				</div>
+
+				<hr class=" border-gray-50 dark:border-gray-850 my-2" />
 
 				<div class="my-3 flex w-full items-center justify-between pr-2">
 					<div class=" self-center text-xs font-medium">
@@ -91,7 +140,7 @@
 					<Switch bind:state={adminConfig.ENABLE_MESSAGE_RATING} />
 				</div>
 
-				<hr class=" dark:border-gray-850 my-2" />
+				<hr class=" border-gray-50 dark:border-gray-850 my-2" />
 
 				<div class=" w-full justify-between">
 					<div class="flex w-full justify-between">
@@ -115,7 +164,7 @@
 					</div>
 				</div>
 
-				<hr class=" dark:border-gray-850 my-2" />
+				<hr class=" border-gray-50 dark:border-gray-850 my-2" />
 
 				<div class=" w-full justify-between">
 					<div class="flex w-full justify-between">
@@ -133,6 +182,196 @@
 				</div>
 			</div>
 		{/if}
+
+		<hr class=" border-gray-50 dark:border-gray-850" />
+
+		<div class=" space-y-3">
+			<div class="mt-2 space-y-2 pr-1.5">
+				<div class="flex justify-between items-center text-sm">
+					<div class="  font-medium">{$i18n.t('LDAP')}</div>
+
+					<div class="mt-1">
+						<Switch
+							bind:state={ENABLE_LDAP}
+							on:change={async () => {
+								updateLdapConfig(localStorage.token, ENABLE_LDAP);
+							}}
+						/>
+					</div>
+				</div>
+
+				{#if ENABLE_LDAP}
+					<div class="flex flex-col gap-1">
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Label')}
+								</div>
+								<input
+									class="w-full bg-transparent outline-none py-0.5"
+									required
+									placeholder={$i18n.t('Enter server label')}
+									bind:value={LDAP_SERVER.label}
+								/>
+							</div>
+							<div class="w-full"></div>
+						</div>
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Host')}
+								</div>
+								<input
+									class="w-full bg-transparent outline-none py-0.5"
+									required
+									placeholder={$i18n.t('Enter server host')}
+									bind:value={LDAP_SERVER.host}
+								/>
+							</div>
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Port')}
+								</div>
+								<Tooltip
+									placement="top-start"
+									content={$i18n.t('Default to 389 or 636 if TLS is enabled')}
+									className="w-full"
+								>
+									<input
+										class="w-full bg-transparent outline-none py-0.5"
+										type="number"
+										placeholder={$i18n.t('Enter server port')}
+										bind:value={LDAP_SERVER.port}
+									/>
+								</Tooltip>
+							</div>
+						</div>
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Application DN')}
+								</div>
+								<Tooltip
+									content={$i18n.t('The Application Account DN you bind with for search')}
+									placement="top-start"
+								>
+									<input
+										class="w-full bg-transparent outline-none py-0.5"
+										required
+										placeholder={$i18n.t('Enter Application DN')}
+										bind:value={LDAP_SERVER.app_dn}
+									/>
+								</Tooltip>
+							</div>
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Application DN Password')}
+								</div>
+								<SensitiveInput
+									placeholder={$i18n.t('Enter Application DN Password')}
+									bind:value={LDAP_SERVER.app_dn_password}
+								/>
+							</div>
+						</div>
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Attribute for Username')}
+								</div>
+								<Tooltip
+									content={$i18n.t(
+										'The LDAP attribute that maps to the username that users use to sign in.'
+									)}
+									placement="top-start"
+								>
+									<input
+										class="w-full bg-transparent outline-none py-0.5"
+										required
+										placeholder={$i18n.t('Example: sAMAccountName or uid or userPrincipalName')}
+										bind:value={LDAP_SERVER.attribute_for_username}
+									/>
+								</Tooltip>
+							</div>
+						</div>
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Search Base')}
+								</div>
+								<Tooltip content={$i18n.t('The base to search for users')} placement="top-start">
+									<input
+										class="w-full bg-transparent outline-none py-0.5"
+										required
+										placeholder={$i18n.t('Example: ou=users,dc=foo,dc=example')}
+										bind:value={LDAP_SERVER.search_base}
+									/>
+								</Tooltip>
+							</div>
+						</div>
+						<div class="flex w-full gap-2">
+							<div class="w-full">
+								<div class=" self-center text-xs font-medium min-w-fit mb-1">
+									{$i18n.t('Search Filters')}
+								</div>
+								<input
+									class="w-full bg-transparent outline-none py-0.5"
+									placeholder={$i18n.t('Example: (&(objectClass=inetOrgPerson)(uid=%s))')}
+									bind:value={LDAP_SERVER.search_filters}
+								/>
+							</div>
+						</div>
+						<div class="text-xs text-gray-400 dark:text-gray-500">
+							<a
+								class=" text-gray-300 font-medium underline"
+								href="https://ldap.com/ldap-filters/"
+								target="_blank"
+							>
+								{$i18n.t('Click here for filter guides.')}
+							</a>
+						</div>
+						<div>
+							<div class="flex justify-between items-center text-sm">
+								<div class="  font-medium">{$i18n.t('TLS')}</div>
+
+								<div class="mt-1">
+									<Switch bind:state={LDAP_SERVER.use_tls} />
+								</div>
+							</div>
+							{#if LDAP_SERVER.use_tls}
+								<div class="flex w-full gap-2">
+									<div class="w-full">
+										<div class=" self-center text-xs font-medium min-w-fit mb-1 mt-1">
+											{$i18n.t('Certificate Path')}
+										</div>
+										<input
+											class="w-full bg-transparent outline-none py-0.5"
+											required
+											placeholder={$i18n.t('Enter certificate path')}
+											bind:value={LDAP_SERVER.certificate_path}
+										/>
+									</div>
+								</div>
+								<div class="flex w-full gap-2">
+									<div class="w-full">
+										<div class=" self-center text-xs font-medium min-w-fit mb-1">
+											{$i18n.t('Ciphers')}
+										</div>
+										<Tooltip content={$i18n.t('Default to ALL')} placement="top-start">
+											<input
+												class="w-full bg-transparent outline-none py-0.5"
+												placeholder={$i18n.t('Example: ALL')}
+												bind:value={LDAP_SERVER.ciphers}
+											/>
+										</Tooltip>
+									</div>
+									<div class="w-full"></div>
+								</div>
+							{/if}
+						</div>
+					</div>
+				{/if}
+			</div>
+		</div>
 	</div>
 
 	<div class="flex justify-end pt-3 text-sm font-medium">

src/lib/components/admin/Settings/Images.svelte

@@ -566,7 +566,7 @@
 
 						<div class="flex gap-2 mb-1">
 							<input
-								class="flex-1 w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+								class="flex-1 w-full text-sm bg-transparent outline-none"
 								placeholder={$i18n.t('API Base URL')}
 								bind:value={config.openai.OPENAI_API_BASE_URL}
 								required

src/lib/components/admin/Settings/Interface.svelte

@@ -25,8 +25,10 @@
 		TASK_MODEL_EXTERNAL: '',
 		TITLE_GENERATION_PROMPT_TEMPLATE: '',
 		TAGS_GENERATION_PROMPT_TEMPLATE: '',
-		ENABLE_SEARCH_QUERY: true,
-		SEARCH_QUERY_GENERATION_PROMPT_TEMPLATE: ''
+		ENABLE_TAGS_GENERATION: true,
+		ENABLE_SEARCH_QUERY_GENERATION: true,
+		ENABLE_RETRIEVAL_QUERY_GENERATION: true,
+		QUERY_GENERATION_PROMPT_TEMPLATE: ''
 	};
 
 	let promptSuggestions = [];
@@ -133,40 +135,26 @@
 				</Tooltip>
 			</div>
 
-			<div class="mt-3">
-				<div class=" mb-2.5 text-xs font-medium">{$i18n.t('Tags Generation Prompt')}</div>
-
-				<Tooltip
-					content={$i18n.t('Leave empty to use the default prompt, or enter a custom prompt')}
-					placement="top-start"
-				>
-					<Textarea
-						bind:value={taskConfig.TAGS_GENERATION_PROMPT_TEMPLATE}
-						placeholder={$i18n.t('Leave empty to use the default prompt, or enter a custom prompt')}
-					/>
-				</Tooltip>
-			</div>
-
 			<hr class=" dark:border-gray-850 my-3" />
 
 			<div class="my-3 flex w-full items-center justify-between">
 				<div class=" self-center text-xs font-medium">
-					{$i18n.t('Enable Web Search Query Generation')}
+					{$i18n.t('Enable Tags Generation')}
 				</div>
 
-				<Switch bind:state={taskConfig.ENABLE_SEARCH_QUERY} />
+				<Switch bind:state={taskConfig.ENABLE_TAGS_GENERATION} />
 			</div>
 
-			{#if taskConfig.ENABLE_SEARCH_QUERY}
-				<div class="">
-					<div class=" mb-2.5 text-xs font-medium">{$i18n.t('Search Query Generation Prompt')}</div>
+			{#if taskConfig.ENABLE_TAGS_GENERATION}
+				<div class="mt-3">
+					<div class=" mb-2.5 text-xs font-medium">{$i18n.t('Tags Generation Prompt')}</div>
 
 					<Tooltip
 						content={$i18n.t('Leave empty to use the default prompt, or enter a custom prompt')}
 						placement="top-start"
 					>
 						<Textarea
-							bind:value={taskConfig.SEARCH_QUERY_GENERATION_PROMPT_TEMPLATE}
+							bind:value={taskConfig.TAGS_GENERATION_PROMPT_TEMPLATE}
 							placeholder={$i18n.t(
 								'Leave empty to use the default prompt, or enter a custom prompt'
 							)}
@@ -174,6 +162,38 @@
 					</Tooltip>
 				</div>
 			{/if}
+
+			<hr class=" dark:border-gray-850 my-3" />
+
+			<div class="my-3 flex w-full items-center justify-between">
+				<div class=" self-center text-xs font-medium">
+					{$i18n.t('Enable Retrieval Query Generation')}
+				</div>
+
+				<Switch bind:state={taskConfig.ENABLE_RETRIEVAL_QUERY_GENERATION} />
+			</div>
+
+			<div class="my-3 flex w-full items-center justify-between">
+				<div class=" self-center text-xs font-medium">
+					{$i18n.t('Enable Web Search Query Generation')}
+				</div>
+
+				<Switch bind:state={taskConfig.ENABLE_SEARCH_QUERY_GENERATION} />
+			</div>
+
+			<div class="">
+				<div class=" mb-2.5 text-xs font-medium">{$i18n.t('Query Generation Prompt')}</div>
+
+				<Tooltip
+					content={$i18n.t('Leave empty to use the default prompt, or enter a custom prompt')}
+					placement="top-start"
+				>
+					<Textarea
+						bind:value={taskConfig.QUERY_GENERATION_PROMPT_TEMPLATE}
+						placeholder={$i18n.t('Leave empty to use the default prompt, or enter a custom prompt')}
+					/>
+				</Tooltip>
+			</div>
 		</div>
 
 		<hr class=" dark:border-gray-850 my-3" />

src/lib/components/admin/Settings/Models.svelte

@@ -1,1090 +1,401 @@
 <script lang="ts">
-	import { toast } from 'svelte-sonner';
-	import { onMount, getContext } from 'svelte';
+	import { marked } from 'marked';
+	import fileSaver from 'file-saver';
+	const { saveAs } = fileSaver;
 
-	import { WEBUI_API_BASE_URL, WEBUI_BASE_URL } from '$lib/constants';
-	import { WEBUI_NAME, models, MODEL_DOWNLOAD_POOL, user, config } from '$lib/stores';
-	import { splitStream } from '$lib/utils';
+	import { onMount, getContext, tick } from 'svelte';
+	const i18n = getContext('i18n');
 
+	import { WEBUI_NAME, config, mobile, models as _models, settings, user } from '$lib/stores';
 	import {
-		createModel,
-		deleteModel,
-		downloadModel,
-		getOllamaUrls,
-		getOllamaVersion,
-		pullModel,
-		uploadModel,
-		getOllamaConfig
-	} from '$lib/apis/ollama';
-	import { getModels as _getModels } from '$lib/apis';
-
+		createNewModel,
+		deleteAllModels,
+		getBaseModels,
+		toggleModelById,
+		updateModelById
+	} from '$lib/apis/models';
+
+	import { getModels } from '$lib/apis';
+	import Search from '$lib/components/icons/Search.svelte';
 	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import Switch from '$lib/components/common/Switch.svelte';
 	import Spinner from '$lib/components/common/Spinner.svelte';
-	import ModelDeleteConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
-
-	const i18n = getContext('i18n');
-
-	const getModels = async () => {
-		return await _getModels(localStorage.token);
-	};
-
-	let modelUploadInputElement: HTMLInputElement;
-
-	let showModelDeleteConfirm = false;
-
-	// Models
-
-	let ollamaEnabled = null;
-
-	let OLLAMA_URLS = [];
-	let selectedOllamaUrlIdx: number | null = null;
-
-	let updateModelId = null;
-	let updateProgress = null;
 
-	let showExperimentalOllama = false;
-
-	let ollamaVersion = null;
-	const MAX_PARALLEL_DOWNLOADS = 3;
-
-	let modelTransferring = false;
-	let modelTag = '';
-
-	let createModelLoading = false;
-	let createModelTag = '';
-	let createModelContent = '';
-	let createModelDigest = '';
-	let createModelPullProgress = null;
-
-	let digest = '';
-	let pullProgress = null;
-
-	let modelUploadMode = 'file';
-	let modelInputFile: File[] | null = null;
-	let modelFileUrl = '';
-	let modelFileContent = `TEMPLATE """{{ .System }}\nUSER: {{ .Prompt }}\nASSISTANT: """\nPARAMETER num_ctx 4096\nPARAMETER stop "</s>"\nPARAMETER stop "USER:"\nPARAMETER stop "ASSISTANT:"`;
-	let modelFileDigest = '';
-
-	let uploadProgress = null;
-	let uploadMessage = '';
-
-	let deleteModelTag = '';
-
-	const updateModelsHandler = async () => {
-		for (const model of $models.filter(
-			(m) =>
-				!(m?.preset ?? false) &&
-				m.owned_by === 'ollama' &&
-				(selectedOllamaUrlIdx === null
-					? true
-					: (m?.ollama?.urls ?? []).includes(selectedOllamaUrlIdx))
-		)) {
-			console.log(model);
-
-			updateModelId = model.id;
-			const [res, controller] = await pullModel(
-				localStorage.token,
-				model.id,
-				selectedOllamaUrlIdx
-			).catch((error) => {
-				toast.error(error);
-				return null;
-			});
-
-			if (res) {
-				const reader = res.body
-					.pipeThrough(new TextDecoderStream())
-					.pipeThrough(splitStream('\n'))
-					.getReader();
+	import ModelEditor from '$lib/components/workspace/Models/ModelEditor.svelte';
+	import { toast } from 'svelte-sonner';
+	import ConfirmDialog from '$lib/components/common/ConfirmDialog.svelte';
 
-				while (true) {
-					try {
-						const { value, done } = await reader.read();
-						if (done) break;
+	let importFiles;
+	let modelsImportInputElement: HTMLInputElement;
 
-						let lines = value.split('\n');
+	let models = null;
 
-						for (const line of lines) {
-							if (line !== '') {
-								let data = JSON.parse(line);
+	let workspaceModels = null;
+	let baseModels = null;
 
-								console.log(data);
-								if (data.error) {
-									throw data.error;
-								}
-								if (data.detail) {
-									throw data.detail;
-								}
-								if (data.status) {
-									if (data.digest) {
-										updateProgress = 0;
-										if (data.completed) {
-											updateProgress = Math.round((data.completed / data.total) * 1000) / 10;
-										} else {
-											updateProgress = 100;
-										}
-									} else {
-										toast.success(data.status);
-									}
-								}
-							}
-						}
-					} catch (error) {
-						console.log(error);
-					}
-				}
-			}
-		}
+	let filteredModels = [];
+	let selectedModelId = null;
+	let showResetModal = false;
 
-		updateModelId = null;
-		updateProgress = null;
-	};
+	$: if (models) {
+		filteredModels = models.filter(
+			(m) => searchValue === '' || m.name.toLowerCase().includes(searchValue.toLowerCase())
+		);
+	}
 
-	const pullModelHandler = async () => {
-		const sanitizedModelTag = modelTag.trim().replace(/^ollama\s+(run|pull)\s+/, '');
-		console.log($MODEL_DOWNLOAD_POOL);
-		if ($MODEL_DOWNLOAD_POOL[sanitizedModelTag]) {
-			toast.error(
-				$i18n.t(`Model '{{modelTag}}' is already in queue for downloading.`, {
-					modelTag: sanitizedModelTag
-				})
-			);
-			return;
-		}
-		if (Object.keys($MODEL_DOWNLOAD_POOL).length === MAX_PARALLEL_DOWNLOADS) {
-			toast.error(
-				$i18n.t('Maximum of 3 models can be downloaded simultaneously. Please try again later.')
-			);
-			return;
-		}
+	let searchValue = '';
 
-		const [res, controller] = await pullModel(
-			localStorage.token,
-			sanitizedModelTag,
-			selectedOllamaUrlIdx
-		).catch((error) => {
-			toast.error(error);
-			return null;
+	const downloadModels = async (models) => {
+		let blob = new Blob([JSON.stringify(models)], {
+			type: 'application/json'
 		});
+		saveAs(blob, `models-export-${Date.now()}.json`);
+	};
 
-		if (res) {
-			const reader = res.body
-				.pipeThrough(new TextDecoderStream())
-				.pipeThrough(splitStream('\n'))
-				.getReader();
-
-			MODEL_DOWNLOAD_POOL.set({
-				...$MODEL_DOWNLOAD_POOL,
-				[sanitizedModelTag]: {
-					...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
-					abortController: controller,
-					reader,
-					done: false
-				}
-			});
-
-			while (true) {
-				try {
-					const { value, done } = await reader.read();
-					if (done) break;
-
-					let lines = value.split('\n');
-
-					for (const line of lines) {
-						if (line !== '') {
-							let data = JSON.parse(line);
-							console.log(data);
-							if (data.error) {
-								throw data.error;
-							}
-							if (data.detail) {
-								throw data.detail;
-							}
-
-							if (data.status) {
-								if (data.digest) {
-									let downloadProgress = 0;
-									if (data.completed) {
-										downloadProgress = Math.round((data.completed / data.total) * 1000) / 10;
-									} else {
-										downloadProgress = 100;
-									}
-
-									MODEL_DOWNLOAD_POOL.set({
-										...$MODEL_DOWNLOAD_POOL,
-										[sanitizedModelTag]: {
-											...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
-											pullProgress: downloadProgress,
-											digest: data.digest
-										}
-									});
-								} else {
-									toast.success(data.status);
-
-									MODEL_DOWNLOAD_POOL.set({
-										...$MODEL_DOWNLOAD_POOL,
-										[sanitizedModelTag]: {
-											...$MODEL_DOWNLOAD_POOL[sanitizedModelTag],
-											done: data.status === 'success'
-										}
-									});
-								}
-							}
-						}
-					}
-				} catch (error) {
-					console.log(error);
-					if (typeof error !== 'string') {
-						error = error.message;
-					}
-
-					toast.error(error);
-					// opts.callback({ success: false, error, modelName: opts.modelName });
-				}
-			}
-
-			console.log($MODEL_DOWNLOAD_POOL[sanitizedModelTag]);
+	const init = async () => {
+		workspaceModels = await getBaseModels(localStorage.token);
+		baseModels = await getModels(localStorage.token, true);
 
-			if ($MODEL_DOWNLOAD_POOL[sanitizedModelTag].done) {
-				toast.success(
-					$i18n.t(`Model '{{modelName}}' has been successfully downloaded.`, {
-						modelName: sanitizedModelTag
-					})
-				);
+		models = baseModels.map((m) => {
+			const workspaceModel = workspaceModels.find((wm) => wm.id === m.id);
 
-				models.set(await getModels());
+			if (workspaceModel) {
+				return {
+					...m,
+					...workspaceModel
+				};
 			} else {
-				toast.error($i18n.t('Download canceled'));
-			}
+				return {
+					...m,
+					id: m.id,
+					name: m.name,
 
-			delete $MODEL_DOWNLOAD_POOL[sanitizedModelTag];
-
-			MODEL_DOWNLOAD_POOL.set({
-				...$MODEL_DOWNLOAD_POOL
-			});
-		}
-
-		modelTag = '';
-		modelTransferring = false;
+					is_active: true
+				};
+			}
+		});
 	};
 
-	const uploadModelHandler = async () => {
-		modelTransferring = true;
+	const upsertModelHandler = async (model) => {
+		model.base_model_id = null;
 
-		let uploaded = false;
-		let fileResponse = null;
-		let name = '';
-
-		if (modelUploadMode === 'file') {
-			const file = modelInputFile ? modelInputFile[0] : null;
-
-			if (file) {
-				uploadMessage = 'Uploading...';
-
-				fileResponse = await uploadModel(localStorage.token, file, selectedOllamaUrlIdx).catch(
-					(error) => {
-						toast.error(error);
-						return null;
-					}
-				);
-			}
-		} else {
-			uploadProgress = 0;
-			fileResponse = await downloadModel(
-				localStorage.token,
-				modelFileUrl,
-				selectedOllamaUrlIdx
-			).catch((error) => {
-				toast.error(error);
+		if (workspaceModels.find((m) => m.id === model.id)) {
+			const res = await updateModelById(localStorage.token, model.id, model).catch((error) => {
 				return null;
 			});
-		}
-
-		if (fileResponse && fileResponse.ok) {
-			const reader = fileResponse.body
-				.pipeThrough(new TextDecoderStream())
-				.pipeThrough(splitStream('\n'))
-				.getReader();
-
-			while (true) {
-				const { value, done } = await reader.read();
-				if (done) break;
-
-				try {
-					let lines = value.split('\n');
-
-					for (const line of lines) {
-						if (line !== '') {
-							let data = JSON.parse(line.replace(/^data: /, ''));
-
-							if (data.progress) {
-								if (uploadMessage) {
-									uploadMessage = '';
-								}
-								uploadProgress = data.progress;
-							}
-
-							if (data.error) {
-								throw data.error;
-							}
 
-							if (data.done) {
-								modelFileDigest = data.blob;
-								name = data.name;
-								uploaded = true;
-							}
-						}
-					}
-				} catch (error) {
-					console.log(error);
-				}
+			if (res) {
+				toast.success($i18n.t('Model updated successfully'));
 			}
 		} else {
-			const error = await fileResponse?.json();
-			toast.error(error?.detail ?? error);
-		}
-
-		if (uploaded) {
-			const res = await createModel(
-				localStorage.token,
-				`${name}:latest`,
-				`FROM @${modelFileDigest}\n${modelFileContent}`
-			);
-
-			if (res && res.ok) {
-				const reader = res.body
-					.pipeThrough(new TextDecoderStream())
-					.pipeThrough(splitStream('\n'))
-					.getReader();
-
-				while (true) {
-					const { value, done } = await reader.read();
-					if (done) break;
-
-					try {
-						let lines = value.split('\n');
-
-						for (const line of lines) {
-							if (line !== '') {
-								console.log(line);
-								let data = JSON.parse(line);
-								console.log(data);
-
-								if (data.error) {
-									throw data.error;
-								}
-								if (data.detail) {
-									throw data.detail;
-								}
-
-								if (data.status) {
-									if (
-										!data.digest &&
-										!data.status.includes('writing') &&
-										!data.status.includes('sha256')
-									) {
-										toast.success(data.status);
-									} else {
-										if (data.digest) {
-											digest = data.digest;
-
-											if (data.completed) {
-												pullProgress = Math.round((data.completed / data.total) * 1000) / 10;
-											} else {
-												pullProgress = 100;
-											}
-										}
-									}
-								}
-							}
-						}
-					} catch (error) {
-						console.log(error);
-						toast.error(error);
-					}
-				}
-			}
-		}
-
-		modelFileUrl = '';
-
-		if (modelUploadInputElement) {
-			modelUploadInputElement.value = '';
-		}
-		modelInputFile = null;
-		modelTransferring = false;
-		uploadProgress = null;
-
-		models.set(await getModels());
-	};
+			const res = await createNewModel(localStorage.token, model).catch((error) => {
+				return null;
+			});
 
-	const deleteModelHandler = async () => {
-		const res = await deleteModel(localStorage.token, deleteModelTag, selectedOllamaUrlIdx).catch(
-			(error) => {
-				toast.error(error);
+			if (res) {
+				toast.success($i18n.t('Model updated successfully'));
 			}
-		);
-
-		if (res) {
-			toast.success($i18n.t(`Deleted {{deleteModelTag}}`, { deleteModelTag }));
 		}
 
-		deleteModelTag = '';
-		models.set(await getModels());
+		_models.set(await getModels(localStorage.token));
+		await init();
 	};
 
-	const cancelModelPullHandler = async (model: string) => {
-		const { reader, abortController } = $MODEL_DOWNLOAD_POOL[model];
-		if (abortController) {
-			abortController.abort();
-		}
-		if (reader) {
-			await reader.cancel();
-			delete $MODEL_DOWNLOAD_POOL[model];
-			MODEL_DOWNLOAD_POOL.set({
-				...$MODEL_DOWNLOAD_POOL
+	const toggleModelHandler = async (model) => {
+		if (!Object.keys(model).includes('base_model_id')) {
+			await createNewModel(localStorage.token, {
+				id: model.id,
+				name: model.name,
+				base_model_id: null,
+				meta: {},
+				params: {},
+				access_control: {},
+				is_active: model.is_active
+			}).catch((error) => {
+				return null;
 			});
-			await deleteModel(localStorage.token, model);
-			toast.success(`${model} download has been canceled`);
-		}
-	};
-
-	const createModelHandler = async () => {
-		createModelLoading = true;
-		const res = await createModel(
-			localStorage.token,
-			createModelTag,
-			createModelContent,
-			selectedOllamaUrlIdx
-		).catch((error) => {
-			toast.error(error);
-			return null;
-		});
-
-		if (res && res.ok) {
-			const reader = res.body
-				.pipeThrough(new TextDecoderStream())
-				.pipeThrough(splitStream('\n'))
-				.getReader();
-
-			while (true) {
-				const { value, done } = await reader.read();
-				if (done) break;
 
-				try {
-					let lines = value.split('\n');
-
-					for (const line of lines) {
-						if (line !== '') {
-							console.log(line);
-							let data = JSON.parse(line);
-							console.log(data);
-
-							if (data.error) {
-								throw data.error;
-							}
-							if (data.detail) {
-								throw data.detail;
-							}
-
-							if (data.status) {
-								if (
-									!data.digest &&
-									!data.status.includes('writing') &&
-									!data.status.includes('sha256')
-								) {
-									toast.success(data.status);
-								} else {
-									if (data.digest) {
-										createModelDigest = data.digest;
-
-										if (data.completed) {
-											createModelPullProgress =
-												Math.round((data.completed / data.total) * 1000) / 10;
-										} else {
-											createModelPullProgress = 100;
-										}
-									}
-								}
-							}
-						}
-					}
-				} catch (error) {
-					console.log(error);
-					toast.error(error);
-				}
-			}
+			await init();
+		} else {
+			await toggleModelById(localStorage.token, model.id);
 		}
 
-		models.set(await getModels());
-
-		createModelLoading = false;
-
-		createModelTag = '';
-		createModelContent = '';
-		createModelDigest = '';
-		createModelPullProgress = null;
+		_models.set(await getModels(localStorage.token));
 	};
 
 	onMount(async () => {
-		const ollamaConfig = await getOllamaConfig(localStorage.token);
-
-		if (ollamaConfig.ENABLE_OLLAMA_API) {
-			ollamaEnabled = true;
-
-			await Promise.all([
-				(async () => {
-					OLLAMA_URLS = await getOllamaUrls(localStorage.token).catch((error) => {
-						toast.error(error);
-						return [];
-					});
-
-					if (OLLAMA_URLS.length > 0) {
-						selectedOllamaUrlIdx = 0;
-					}
-				})(),
-				(async () => {
-					ollamaVersion = await getOllamaVersion(localStorage.token).catch((error) => false);
-				})()
-			]);
-		} else {
-			ollamaEnabled = false;
-			toast.error($i18n.t('Ollama API is disabled'));
-		}
+		init();
 	});
 </script>
 
-<ModelDeleteConfirmDialog
-	bind:show={showModelDeleteConfirm}
-	on:confirm={() => {
-		deleteModelHandler();
+<ConfirmDialog
+	title={$i18n.t('Delete All Models')}
+	message={$i18n.t('This will delete all models including custom models and cannot be undone.')}
+	bind:show={showResetModal}
+	onConfirm={async () => {
+		const res = deleteAllModels(localStorage.token);
+		if (res) {
+			toast.success($i18n.t('All models deleted successfully'));
+			init();
+		}
 	}}
 />
 
-<div class="flex flex-col h-full justify-between text-sm">
-	<div class=" space-y-3 overflow-y-scroll scrollbar-hidden h-full">
-		{#if ollamaEnabled}
-			{#if ollamaVersion !== null}
-				<div class="space-y-2 pr-1.5">
-					<div class="text-sm font-medium">{$i18n.t('Manage Ollama Models')}</div>
-
-					{#if OLLAMA_URLS.length > 0}
-						<div class="flex gap-2">
-							<div class="flex-1 pb-1">
-								<select
-									class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-									bind:value={selectedOllamaUrlIdx}
-									placeholder={$i18n.t('Select an Ollama instance')}
-								>
-									{#each OLLAMA_URLS as url, idx}
-										<option value={idx} class="bg-gray-50 dark:bg-gray-700">{url}</option>
-									{/each}
-								</select>
-							</div>
+{#if models !== null}
+	{#if selectedModelId === null}
+		<div class="flex flex-col gap-1 mt-1.5 mb-2">
+			<div class="flex justify-between items-center">
+				<div class="flex items-center md:self-center text-xl font-medium px-0.5">
+					{$i18n.t('Models')}
+					<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
+					<span class="text-lg font-medium text-gray-500 dark:text-gray-300"
+						>{filteredModels.length}</span
+					>
+				</div>
 
-							<div>
-								<div class="flex w-full justify-end">
-									<Tooltip content="Update All Models" placement="top">
-										<button
-											class="p-2.5 flex gap-2 items-center bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
-											on:click={() => {
-												updateModelsHandler();
-											}}
-										>
-											<svg
-												xmlns="http://www.w3.org/2000/svg"
-												viewBox="0 0 16 16"
-												fill="currentColor"
-												class="w-4 h-4"
-											>
-												<path
-													d="M7 1a.75.75 0 0 1 .75.75V6h-1.5V1.75A.75.75 0 0 1 7 1ZM6.25 6v2.94L5.03 7.72a.75.75 0 0 0-1.06 1.06l2.5 2.5a.75.75 0 0 0 1.06 0l2.5-2.5a.75.75 0 1 0-1.06-1.06L7.75 8.94V6H10a2 2 0 0 1 2 2v3a2 2 0 0 1-2 2H4a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h2.25Z"
-												/>
-												<path
-													d="M4.268 14A2 2 0 0 0 6 15h6a2 2 0 0 0 2-2v-3a2 2 0 0 0-1-1.732V11a3 3 0 0 1-3 3H4.268Z"
-												/>
-											</svg>
-										</button>
-									</Tooltip>
-								</div>
+				<div>
+					<Tooltip content={$i18n.t('This will delete all models including custom models')}>
+						<button
+							class=" px-2.5 py-1 rounded-full flex gap-1 items-center"
+							type="button"
+							on:click={() => {
+								showResetModal = true;
+							}}
+						>
+							<div class="text-xs flex-shrink-0">
+								{$i18n.t('Reset')}
 							</div>
-						</div>
-
-						{#if updateModelId}
-							Updating "{updateModelId}" {updateProgress ? `(${updateProgress}%)` : ''}
-						{/if}
-					{/if}
+						</button>
+					</Tooltip>
+				</div>
+			</div>
 
-					<div class="space-y-2">
-						<div>
-							<div class=" mb-2 text-sm font-medium">{$i18n.t('Pull a model from Ollama.com')}</div>
-							<div class="flex w-full">
-								<div class="flex-1 mr-2">
-									<input
-										class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-										placeholder={$i18n.t('Enter model tag (e.g. {{modelTag}})', {
-											modelTag: 'mistral:7b'
-										})}
-										bind:value={modelTag}
+			<div class=" flex flex-1 items-center w-full space-x-2">
+				<div class="flex flex-1 items-center">
+					<div class=" self-center ml-1 mr-3">
+						<Search className="size-3.5" />
+					</div>
+					<input
+						class=" w-full text-sm py-1 rounded-r-xl outline-none bg-transparent"
+						bind:value={searchValue}
+						placeholder={$i18n.t('Search Models')}
+					/>
+				</div>
+			</div>
+		</div>
+
+		<div class=" my-2 mb-5" id="model-list">
+			{#if models.length > 0}
+				{#each filteredModels as model (model.id)}
+					<div
+						class=" flex space-x-4 cursor-pointer w-full px-3 py-2 dark:hover:bg-white/5 hover:bg-black/5 rounded-lg transition"
+						id="model-item-{model.id}"
+					>
+						<button
+							class=" flex flex-1 text-left space-x-3.5 cursor-pointer w-full"
+							type="button"
+							on:click={() => {
+								selectedModelId = model.id;
+							}}
+						>
+							<div class=" self-center w-8">
+								<div
+									class=" rounded-full object-cover {(model?.is_active ?? true)
+										? ''
+										: 'opacity-50 dark:opacity-50'} "
+								>
+									<img
+										src={model?.meta?.profile_image_url ?? '/static/favicon.png'}
+										alt="modelfile profile"
+										class=" rounded-full w-full h-auto object-cover"
 									/>
 								</div>
-								<button
-									class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
-									on:click={() => {
-										pullModelHandler();
-									}}
-									disabled={modelTransferring}
-								>
-									{#if modelTransferring}
-										<div class="self-center">
-											<svg
-												class=" w-4 h-4"
-												viewBox="0 0 24 24"
-												fill="currentColor"
-												xmlns="http://www.w3.org/2000/svg"
-											>
-												<style>
-													.spinner_ajPY {
-														transform-origin: center;
-														animation: spinner_AtaB 0.75s infinite linear;
-													}
-
-													@keyframes spinner_AtaB {
-														100% {
-															transform: rotate(360deg);
-														}
-													}
-												</style>
-												<path
-													d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
-													opacity=".25"
-												/>
-												<path
-													d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
-													class="spinner_ajPY"
-												/>
-											</svg>
-										</div>
-									{:else}
-										<svg
-											xmlns="http://www.w3.org/2000/svg"
-											viewBox="0 0 16 16"
-											fill="currentColor"
-											class="w-4 h-4"
-										>
-											<path
-												d="M8.75 2.75a.75.75 0 0 0-1.5 0v5.69L5.03 6.22a.75.75 0 0 0-1.06 1.06l3.5 3.5a.75.75 0 0 0 1.06 0l3.5-3.5a.75.75 0 0 0-1.06-1.06L8.75 8.44V2.75Z"
-											/>
-											<path
-												d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
-											/>
-										</svg>
-									{/if}
-								</button>
 							</div>
 
-							<div class="mt-2 mb-1 text-xs text-gray-400 dark:text-gray-500">
-								{$i18n.t('To access the available model names for downloading,')}
-								<a
-									class=" text-gray-500 dark:text-gray-300 font-medium underline"
-									href="https://ollama.com/library"
-									target="_blank">{$i18n.t('click here.')}</a
+							<div class=" flex-1 self-center {(model?.is_active ?? true) ? '' : 'text-gray-500'}">
+								<Tooltip
+									content={marked.parse(
+										!!model?.meta?.description
+											? model?.meta?.description
+											: model?.ollama?.digest
+												? `${model?.ollama?.digest} **(${model?.ollama?.modified_at})**`
+												: model.id
+									)}
+									className=" w-fit"
+									placement="top-start"
 								>
-							</div>
-
-							{#if Object.keys($MODEL_DOWNLOAD_POOL).length > 0}
-								{#each Object.keys($MODEL_DOWNLOAD_POOL) as model}
-									{#if 'pullProgress' in $MODEL_DOWNLOAD_POOL[model]}
-										<div class="flex flex-col">
-											<div class="font-medium mb-1">{model}</div>
-											<div class="">
-												<div class="flex flex-row justify-between space-x-4 pr-2">
-													<div class=" flex-1">
-														<div
-															class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
-															style="width: {Math.max(
-																15,
-																$MODEL_DOWNLOAD_POOL[model].pullProgress ?? 0
-															)}%"
-														>
-															{$MODEL_DOWNLOAD_POOL[model].pullProgress ?? 0}%
-														</div>
-													</div>
-
-													<Tooltip content={$i18n.t('Cancel')}>
-														<button
-															class="text-gray-800 dark:text-gray-100"
-															on:click={() => {
-																cancelModelPullHandler(model);
-															}}
-														>
-															<svg
-																class="w-4 h-4 text-gray-800 dark:text-white"
-																aria-hidden="true"
-																xmlns="http://www.w3.org/2000/svg"
-																width="24"
-																height="24"
-																fill="currentColor"
-																viewBox="0 0 24 24"
-															>
-																<path
-																	stroke="currentColor"
-																	stroke-linecap="round"
-																	stroke-linejoin="round"
-																	stroke-width="2"
-																	d="M6 18 17.94 6M18 18 6.06 6"
-																/>
-															</svg>
-														</button>
-													</Tooltip>
-												</div>
-												{#if 'digest' in $MODEL_DOWNLOAD_POOL[model]}
-													<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
-														{$MODEL_DOWNLOAD_POOL[model].digest}
-													</div>
-												{/if}
-											</div>
-										</div>
-									{/if}
-								{/each}
-							{/if}
-						</div>
-
-						<div>
-							<div class=" mb-2 text-sm font-medium">{$i18n.t('Delete a model')}</div>
-							<div class="flex w-full">
-								<div class="flex-1 mr-2">
-									<select
-										class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-										bind:value={deleteModelTag}
-										placeholder={$i18n.t('Select a model')}
-									>
-										{#if !deleteModelTag}
-											<option value="" disabled selected>{$i18n.t('Select a model')}</option>
-										{/if}
-										{#each $models.filter((m) => !(m?.preset ?? false) && m.owned_by === 'ollama' && (selectedOllamaUrlIdx === null ? true : (m?.ollama?.urls ?? []).includes(selectedOllamaUrlIdx))) as model}
-											<option value={model.id} class="bg-gray-50 dark:bg-gray-700"
-												>{model.name +
-													' (' +
-													(model.ollama.size / 1024 ** 3).toFixed(1) +
-													' GB)'}</option
-											>
-										{/each}
-									</select>
+									<div class="  font-semibold line-clamp-1">{model.name}</div>
+								</Tooltip>
+								<div class=" text-xs overflow-hidden text-ellipsis line-clamp-1 text-gray-500">
+									<span class=" line-clamp-1">
+										{!!model?.meta?.description
+											? model?.meta?.description
+											: model?.ollama?.digest
+												? `${model.id} (${model?.ollama?.digest})`
+												: model.id}
+									</span>
 								</div>
-								<button
-									class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition"
-									on:click={() => {
-										showModelDeleteConfirm = true;
-									}}
-								>
-									<svg
-										xmlns="http://www.w3.org/2000/svg"
-										viewBox="0 0 16 16"
-										fill="currentColor"
-										class="w-4 h-4"
-									>
-										<path
-											fill-rule="evenodd"
-											d="M5 3.25V4H2.75a.75.75 0 0 0 0 1.5h.3l.815 8.15A1.5 1.5 0 0 0 5.357 15h5.285a1.5 1.5 0 0 0 1.493-1.35l.815-8.15h.3a.75.75 0 0 0 0-1.5H11v-.75A2.25 2.25 0 0 0 8.75 1h-1.5A2.25 2.25 0 0 0 5 3.25Zm2.25-.75a.75.75 0 0 0-.75.75V4h3v-.75a.75.75 0 0 0-.75-.75h-1.5ZM6.05 6a.75.75 0 0 1 .787.713l.275 5.5a.75.75 0 0 1-1.498.075l-.275-5.5A.75.75 0 0 1 6.05 6Zm3.9 0a.75.75 0 0 1 .712.787l-.275 5.5a.75.75 0 0 1-1.498-.075l.275-5.5a.75.75 0 0 1 .786-.711Z"
-											clip-rule="evenodd"
-										/>
-									</svg>
-								</button>
 							</div>
-						</div>
-
-						<div>
-							<div class=" mb-2 text-sm font-medium">{$i18n.t('Create a model')}</div>
-							<div class="flex w-full">
-								<div class="flex-1 mr-2 flex flex-col gap-2">
-									<input
-										class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-										placeholder={$i18n.t('Enter model tag (e.g. {{modelTag}})', {
-											modelTag: 'my-modelfile'
-										})}
-										bind:value={createModelTag}
-										disabled={createModelLoading}
-									/>
-
-									<textarea
-										bind:value={createModelContent}
-										class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-100 dark:bg-gray-850 outline-none resize-none scrollbar-hidden"
-										rows="6"
-										placeholder={`TEMPLATE """{{ .System }}\nUSER: {{ .Prompt }}\nASSISTANT: """\nPARAMETER num_ctx 4096\nPARAMETER stop "</s>"\nPARAMETER stop "USER:"\nPARAMETER stop "ASSISTANT:"`}
-										disabled={createModelLoading}
+						</button>
+						<div class="flex flex-row gap-0.5 items-center self-center">
+							<button
+								class="self-center w-fit text-sm px-2 py-2 dark:text-gray-300 dark:hover:text-white hover:bg-black/5 dark:hover:bg-white/5 rounded-xl"
+								type="button"
+								on:click={() => {
+									selectedModelId = model.id;
+								}}
+							>
+								<svg
+									xmlns="http://www.w3.org/2000/svg"
+									fill="none"
+									viewBox="0 0 24 24"
+									stroke-width="1.5"
+									stroke="currentColor"
+									class="w-4 h-4"
+								>
+									<path
+										stroke-linecap="round"
+										stroke-linejoin="round"
+										d="m16.862 4.487 1.687-1.688a1.875 1.875 0 1 1 2.652 2.652L6.832 19.82a4.5 4.5 0 0 1-1.897 1.13l-2.685.8.8-2.685a4.5 4.5 0 0 1 1.13-1.897L16.863 4.487Zm0 0L19.5 7.125"
 									/>
-								</div>
-
-								<div class="flex self-start">
-									<button
-										class="px-2.5 py-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg transition disabled:cursor-not-allowed"
-										on:click={() => {
-											createModelHandler();
-										}}
-										disabled={createModelLoading}
-									>
-										<svg
-											xmlns="http://www.w3.org/2000/svg"
-											viewBox="0 0 16 16"
-											fill="currentColor"
-											class="size-4"
-										>
-											<path
-												d="M7.25 10.25a.75.75 0 0 0 1.5 0V4.56l2.22 2.22a.75.75 0 1 0 1.06-1.06l-3.5-3.5a.75.75 0 0 0-1.06 0l-3.5 3.5a.75.75 0 0 0 1.06 1.06l2.22-2.22v5.69Z"
-											/>
-											<path
-												d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
-											/>
-										</svg>
-									</button>
-								</div>
-							</div>
+								</svg>
+							</button>
 
-							{#if createModelDigest !== ''}
-								<div class="flex flex-col mt-1">
-									<div class="font-medium mb-1">{createModelTag}</div>
-									<div class="">
-										<div class="flex flex-row justify-between space-x-4 pr-2">
-											<div class=" flex-1">
-												<div
-													class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
-													style="width: {Math.max(15, createModelPullProgress ?? 0)}%"
-												>
-													{createModelPullProgress ?? 0}%
-												</div>
-											</div>
-										</div>
-										{#if createModelDigest}
-											<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
-												{createModelDigest}
-											</div>
-										{/if}
-									</div>
-								</div>
-							{/if}
-						</div>
-
-						<div class="pt-1">
-							<div class="flex justify-between items-center text-xs">
-								<div class=" text-sm font-medium">{$i18n.t('Experimental')}</div>
-								<button
-									class=" text-xs font-medium text-gray-500"
-									type="button"
-									on:click={() => {
-										showExperimentalOllama = !showExperimentalOllama;
-									}}>{showExperimentalOllama ? $i18n.t('Hide') : $i18n.t('Show')}</button
+							<div class="ml-1">
+								<Tooltip
+									content={(model?.is_active ?? true) ? $i18n.t('Enabled') : $i18n.t('Disabled')}
 								>
+									<Switch
+										bind:state={model.is_active}
+										on:change={async () => {
+											toggleModelHandler(model);
+										}}
+									/>
+								</Tooltip>
 							</div>
 						</div>
-
-						{#if showExperimentalOllama}
-							<form
-								on:submit|preventDefault={() => {
-									uploadModelHandler();
-								}}
-							>
-								<div class=" mb-2 flex w-full justify-between">
-									<div class="  text-sm font-medium">{$i18n.t('Upload a GGUF model')}</div>
-
-									<button
-										class="p-1 px-3 text-xs flex rounded transition"
-										on:click={() => {
-											if (modelUploadMode === 'file') {
-												modelUploadMode = 'url';
-											} else {
-												modelUploadMode = 'file';
-											}
-										}}
-										type="button"
-									>
-										{#if modelUploadMode === 'file'}
-											<span class="ml-2 self-center">{$i18n.t('File Mode')}</span>
-										{:else}
-											<span class="ml-2 self-center">{$i18n.t('URL Mode')}</span>
-										{/if}
-									</button>
-								</div>
-
-								<div class="flex w-full mb-1.5">
-									<div class="flex flex-col w-full">
-										{#if modelUploadMode === 'file'}
-											<div
-												class="flex-1 {modelInputFile && modelInputFile.length > 0 ? 'mr-2' : ''}"
-											>
-												<input
-													id="model-upload-input"
-													bind:this={modelUploadInputElement}
-													type="file"
-													bind:files={modelInputFile}
-													on:change={() => {
-														console.log(modelInputFile);
-													}}
-													accept=".gguf,.safetensors"
-													required
-													hidden
-												/>
-
-												<button
-													type="button"
-													class="w-full rounded-lg text-left py-2 px-4 bg-gray-50 dark:text-gray-300 dark:bg-gray-850"
-													on:click={() => {
-														modelUploadInputElement.click();
-													}}
-												>
-													{#if modelInputFile && modelInputFile.length > 0}
-														{modelInputFile[0].name}
-													{:else}
-														{$i18n.t('Click here to select')}
-													{/if}
-												</button>
-											</div>
-										{:else}
-											<div class="flex-1 {modelFileUrl !== '' ? 'mr-2' : ''}">
-												<input
-													class="w-full rounded-lg text-left py-2 px-4 bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none {modelFileUrl !==
-													''
-														? 'mr-2'
-														: ''}"
-													type="url"
-													required
-													bind:value={modelFileUrl}
-													placeholder={$i18n.t('Type Hugging Face Resolve (Download) URL')}
-												/>
-											</div>
-										{/if}
-									</div>
-
-									{#if (modelUploadMode === 'file' && modelInputFile && modelInputFile.length > 0) || (modelUploadMode === 'url' && modelFileUrl !== '')}
-										<button
-											class="px-2.5 bg-gray-50 hover:bg-gray-200 text-gray-800 dark:bg-gray-850 dark:hover:bg-gray-800 dark:text-gray-100 rounded-lg disabled:cursor-not-allowed transition"
-											type="submit"
-											disabled={modelTransferring}
-										>
-											{#if modelTransferring}
-												<div class="self-center">
-													<svg
-														class=" w-4 h-4"
-														viewBox="0 0 24 24"
-														fill="currentColor"
-														xmlns="http://www.w3.org/2000/svg"
-													>
-														<style>
-															.spinner_ajPY {
-																transform-origin: center;
-																animation: spinner_AtaB 0.75s infinite linear;
-															}
-
-															@keyframes spinner_AtaB {
-																100% {
-																	transform: rotate(360deg);
-																}
-															}
-														</style>
-														<path
-															d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
-															opacity=".25"
-														/>
-														<path
-															d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
-															class="spinner_ajPY"
-														/>
-													</svg>
-												</div>
-											{:else}
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														d="M7.25 10.25a.75.75 0 0 0 1.5 0V4.56l2.22 2.22a.75.75 0 1 0 1.06-1.06l-3.5-3.5a.75.75 0 0 0-1.06 0l-3.5 3.5a.75.75 0 0 0 1.06 1.06l2.22-2.22v5.69Z"
-													/>
-													<path
-														d="M3.5 9.75a.75.75 0 0 0-1.5 0v1.5A2.75 2.75 0 0 0 4.75 14h6.5A2.75 2.75 0 0 0 14 11.25v-1.5a.75.75 0 0 0-1.5 0v1.5c0 .69-.56 1.25-1.25 1.25h-6.5c-.69 0-1.25-.56-1.25-1.25v-1.5Z"
-													/>
-												</svg>
-											{/if}
-										</button>
-									{/if}
-								</div>
-
-								{#if (modelUploadMode === 'file' && modelInputFile && modelInputFile.length > 0) || (modelUploadMode === 'url' && modelFileUrl !== '')}
-									<div>
-										<div>
-											<div class=" my-2.5 text-sm font-medium">{$i18n.t('Modelfile Content')}</div>
-											<textarea
-												bind:value={modelFileContent}
-												class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-100 dark:bg-gray-850 outline-none resize-none"
-												rows="6"
-											/>
-										</div>
-									</div>
-								{/if}
-								<div class=" mt-1 text-xs text-gray-400 dark:text-gray-500">
-									{$i18n.t('To access the GGUF models available for downloading,')}
-									<a
-										class=" text-gray-500 dark:text-gray-300 font-medium underline"
-										href="https://huggingface.co/models?search=gguf"
-										target="_blank">{$i18n.t('click here.')}</a
-									>
-								</div>
-
-								{#if uploadMessage}
-									<div class="mt-2">
-										<div class=" mb-2 text-xs">{$i18n.t('Upload Progress')}</div>
-
-										<div class="w-full rounded-full dark:bg-gray-800">
-											<div
-												class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
-												style="width: 100%"
-											>
-												{uploadMessage}
-											</div>
-										</div>
-										<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
-											{modelFileDigest}
-										</div>
-									</div>
-								{:else if uploadProgress !== null}
-									<div class="mt-2">
-										<div class=" mb-2 text-xs">{$i18n.t('Upload Progress')}</div>
-
-										<div class="w-full rounded-full dark:bg-gray-800">
-											<div
-												class="dark:bg-gray-600 bg-gray-500 text-xs font-medium text-gray-100 text-center p-0.5 leading-none rounded-full"
-												style="width: {Math.max(15, uploadProgress ?? 0)}%"
-											>
-												{uploadProgress ?? 0}%
-											</div>
-										</div>
-										<div class="mt-1 text-xs dark:text-gray-500" style="font-size: 0.5rem;">
-											{modelFileDigest}
-										</div>
-									</div>
-								{/if}
-							</form>
-						{/if}
 					</div>
-				</div>
-			{:else if ollamaVersion === false}
-				<div>Ollama Not Detected</div>
+				{/each}
 			{:else}
-				<div class="flex h-full justify-center">
-					<div class="my-auto">
-						<Spinner className="size-6" />
+				<div class="flex flex-col items-center justify-center w-full h-20">
+					<div class="text-gray-500 dark:text-gray-400 text-xs">
+						{$i18n.t('No models found')}
 					</div>
 				</div>
 			{/if}
-		{:else if ollamaEnabled === false}
-			<div>{$i18n.t('Ollama API is disabled')}</div>
-		{:else}
-			<div class="flex h-full justify-center">
-				<div class="my-auto">
-					<Spinner className="size-6" />
+		</div>
+
+		{#if $user?.role === 'admin'}
+			<div class=" flex justify-end w-full mb-3">
+				<div class="flex space-x-1">
+					<input
+						id="models-import-input"
+						bind:this={modelsImportInputElement}
+						bind:files={importFiles}
+						type="file"
+						accept=".json"
+						hidden
+						on:change={() => {
+							console.log(importFiles);
+
+							let reader = new FileReader();
+							reader.onload = async (event) => {
+								let savedModels = JSON.parse(event.target.result);
+								console.log(savedModels);
+
+								for (const model of savedModels) {
+									if (Object.keys(model).includes('base_model_id')) {
+										if (model.base_model_id === null) {
+											upsertModelHandler(model);
+										}
+									} else {
+										if (model?.info ?? false) {
+											if (model.info.base_model_id === null) {
+												upsertModelHandler(model.info);
+											}
+										}
+									}
+								}
+
+								await _models.set(await getModels(localStorage.token));
+								init();
+							};
+
+							reader.readAsText(importFiles[0]);
+						}}
+					/>
+
+					<button
+						class="flex text-xs items-center space-x-1 px-3 py-1.5 rounded-xl bg-gray-50 hover:bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 dark:text-gray-200 transition"
+						on:click={() => {
+							modelsImportInputElement.click();
+						}}
+					>
+						<div class=" self-center mr-2 font-medium line-clamp-1">
+							{$i18n.t('Import Presets')}
+						</div>
+
+						<div class=" self-center">
+							<svg
+								xmlns="http://www.w3.org/2000/svg"
+								viewBox="0 0 16 16"
+								fill="currentColor"
+								class="w-3.5 h-3.5"
+							>
+								<path
+									fill-rule="evenodd"
+									d="M4 2a1.5 1.5 0 0 0-1.5 1.5v9A1.5 1.5 0 0 0 4 14h8a1.5 1.5 0 0 0 1.5-1.5V6.621a1.5 1.5 0 0 0-.44-1.06L9.94 2.439A1.5 1.5 0 0 0 8.878 2H4Zm4 9.5a.75.75 0 0 1-.75-.75V8.06l-.72.72a.75.75 0 0 1-1.06-1.06l2-2a.75.75 0 0 1 1.06 0l2 2a.75.75 0 1 1-1.06 1.06l-.72-.72v2.69a.75.75 0 0 1-.75.75Z"
+									clip-rule="evenodd"
+								/>
+							</svg>
+						</div>
+					</button>
+
+					<button
+						class="flex text-xs items-center space-x-1 px-3 py-1.5 rounded-xl bg-gray-50 hover:bg-gray-100 dark:bg-gray-800 dark:hover:bg-gray-700 dark:text-gray-200 transition"
+						on:click={async () => {
+							downloadModels(models);
+						}}
+					>
+						<div class=" self-center mr-2 font-medium line-clamp-1">
+							{$i18n.t('Export Presets')}
+						</div>
+
+						<div class=" self-center">
+							<svg
+								xmlns="http://www.w3.org/2000/svg"
+								viewBox="0 0 16 16"
+								fill="currentColor"
+								class="w-3.5 h-3.5"
+							>
+								<path
+									fill-rule="evenodd"
+									d="M4 2a1.5 1.5 0 0 0-1.5 1.5v9A1.5 1.5 0 0 0 4 14h8a1.5 1.5 0 0 0 1.5-1.5V6.621a1.5 1.5 0 0 0-.44-1.06L9.94 2.439A1.5 1.5 0 0 0 8.878 2H4Zm4 3.5a.75.75 0 0 1 .75.75v2.69l.72-.72a.75.75 0 1 1 1.06 1.06l-2 2a.75.75 0 0 1-1.06 0l-2-2a.75.75 0 0 1 1.06-1.06l.72.72V6.25A.75.75 0 0 1 8 5.5Z"
+									clip-rule="evenodd"
+								/>
+							</svg>
+						</div>
+					</button>
 				</div>
 			</div>
 		{/if}
+	{:else}
+		<ModelEditor
+			edit
+			model={models.find((m) => m.id === selectedModelId)}
+			preset={false}
+			onSubmit={(model) => {
+				console.log(model);
+				upsertModelHandler(model);
+				selectedModelId = null;
+			}}
+			onBack={() => {
+				selectedModelId = null;
+			}}
+		/>
+	{/if}
+{:else}
+	<div class=" h-full w-full flex justify-center items-center">
+		<Spinner />
 	</div>
-</div>
+{/if}

src/lib/components/admin/Settings/Users.svelte

@@ -1,214 +0,0 @@
-<script lang="ts">
-	import { getBackendConfig, getModelFilterConfig, updateModelFilterConfig } from '$lib/apis';
-	import { getSignUpEnabledStatus, toggleSignUpEnabledStatus } from '$lib/apis/auths';
-	import { getUserPermissions, updateUserPermissions } from '$lib/apis/users';
-
-	import { onMount, getContext } from 'svelte';
-	import { models, config } from '$lib/stores';
-	import Switch from '$lib/components/common/Switch.svelte';
-	import { setDefaultModels } from '$lib/apis/configs';
-
-	const i18n = getContext('i18n');
-
-	export let saveHandler: Function;
-
-	let defaultModelId = '';
-
-	let whitelistEnabled = false;
-	let whitelistModels = [''];
-	let permissions = {
-		chat: {
-			deletion: true,
-			edit: true,
-			temporary: true
-		}
-	};
-
-	let chatDeletion = true;
-	let chatEdit = true;
-	let chatTemporary = true;
-
-	onMount(async () => {
-		permissions = await getUserPermissions(localStorage.token);
-
-		chatDeletion = permissions?.chat?.deletion ?? true;
-		chatEdit = permissions?.chat?.editing ?? true;
-		chatTemporary = permissions?.chat?.temporary ?? true;
-
-		const res = await getModelFilterConfig(localStorage.token);
-		if (res) {
-			whitelistEnabled = res.enabled;
-			whitelistModels = res.models.length > 0 ? res.models : [''];
-		}
-
-		defaultModelId = $config.default_models ? $config?.default_models.split(',')[0] : '';
-	});
-</script>
-
-<form
-	class="flex flex-col h-full justify-between space-y-3 text-sm"
-	on:submit|preventDefault={async () => {
-		// console.log('submit');
-
-		await setDefaultModels(localStorage.token, defaultModelId);
-		await updateUserPermissions(localStorage.token, {
-			chat: {
-				deletion: chatDeletion,
-				editing: chatEdit,
-				temporary: chatTemporary
-			}
-		});
-		await updateModelFilterConfig(localStorage.token, whitelistEnabled, whitelistModels);
-		saveHandler();
-
-		await config.set(await getBackendConfig());
-	}}
->
-	<div class=" space-y-3 overflow-y-scroll max-h-full">
-		<div>
-			<div class=" mb-2 text-sm font-medium">{$i18n.t('User Permissions')}</div>
-
-			<div class="  flex w-full justify-between my-2 pr-2">
-				<div class=" self-center text-xs font-medium">{$i18n.t('Allow Chat Deletion')}</div>
-
-				<Switch bind:state={chatDeletion} />
-			</div>
-
-			<div class="  flex w-full justify-between my-2 pr-2">
-				<div class=" self-center text-xs font-medium">{$i18n.t('Allow Chat Editing')}</div>
-
-				<Switch bind:state={chatEdit} />
-			</div>
-
-			<div class="  flex w-full justify-between my-2 pr-2">
-				<div class=" self-center text-xs font-medium">{$i18n.t('Allow Temporary Chat')}</div>
-
-				<Switch bind:state={chatTemporary} />
-			</div>
-		</div>
-
-		<hr class=" dark:border-gray-850 my-2" />
-
-		<div class="mt-2 space-y-3">
-			<div>
-				<div class="mb-2">
-					<div class="flex justify-between items-center text-xs">
-						<div class=" text-sm font-medium">{$i18n.t('Manage Models')}</div>
-					</div>
-				</div>
-				<div class=" space-y-1 mb-3">
-					<div class="mb-2">
-						<div class="flex justify-between items-center text-xs">
-							<div class=" text-xs font-medium">{$i18n.t('Default Model')}</div>
-						</div>
-					</div>
-
-					<div class="flex-1 mr-2">
-						<select
-							class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-							bind:value={defaultModelId}
-							placeholder="Select a model"
-						>
-							<option value="" disabled selected>{$i18n.t('Select a model')}</option>
-							{#each $models.filter((model) => model.id) as model}
-								<option value={model.id} class="bg-gray-100 dark:bg-gray-700">{model.name}</option>
-							{/each}
-						</select>
-					</div>
-				</div>
-
-				<div class=" space-y-1">
-					<div class="mb-2">
-						<div class="flex justify-between items-center text-xs my-3 pr-2">
-							<div class=" text-xs font-medium">{$i18n.t('Model Whitelisting')}</div>
-
-							<Switch bind:state={whitelistEnabled} />
-						</div>
-					</div>
-
-					{#if whitelistEnabled}
-						<div>
-							<div class=" space-y-1.5">
-								{#each whitelistModels as modelId, modelIdx}
-									<div class="flex w-full">
-										<div class="flex-1 mr-2">
-											<select
-												class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
-												bind:value={modelId}
-												placeholder="Select a model"
-											>
-												<option value="" disabled selected>{$i18n.t('Select a model')}</option>
-												{#each $models.filter((model) => model.id) as model}
-													<option value={model.id} class="bg-gray-100 dark:bg-gray-700"
-														>{model.name}</option
-													>
-												{/each}
-											</select>
-										</div>
-
-										{#if modelIdx === 0}
-											<button
-												class="px-2.5 bg-gray-100 hover:bg-gray-200 text-gray-800 dark:bg-gray-900 dark:text-white rounded-lg transition"
-												type="button"
-												on:click={() => {
-													if (whitelistModels.at(-1) !== '') {
-														whitelistModels = [...whitelistModels, ''];
-													}
-												}}
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path
-														d="M8.75 3.75a.75.75 0 0 0-1.5 0v3.5h-3.5a.75.75 0 0 0 0 1.5h3.5v3.5a.75.75 0 0 0 1.5 0v-3.5h3.5a.75.75 0 0 0 0-1.5h-3.5v-3.5Z"
-													/>
-												</svg>
-											</button>
-										{:else}
-											<button
-												class="px-2.5 bg-gray-100 hover:bg-gray-200 text-gray-800 dark:bg-gray-900 dark:text-white rounded-lg transition"
-												type="button"
-												on:click={() => {
-													whitelistModels.splice(modelIdx, 1);
-													whitelistModels = whitelistModels;
-												}}
-											>
-												<svg
-													xmlns="http://www.w3.org/2000/svg"
-													viewBox="0 0 16 16"
-													fill="currentColor"
-													class="w-4 h-4"
-												>
-													<path d="M3.75 7.25a.75.75 0 0 0 0 1.5h8.5a.75.75 0 0 0 0-1.5h-8.5Z" />
-												</svg>
-											</button>
-										{/if}
-									</div>
-								{/each}
-							</div>
-
-							<div class="flex justify-end items-center text-xs mt-1.5 text-right">
-								<div class=" text-xs font-medium">
-									{whitelistModels.length}
-									{$i18n.t('Model(s) Whitelisted')}
-								</div>
-							</div>
-						</div>
-					{/if}
-				</div>
-			</div>
-		</div>
-	</div>
-
-	<div class="flex justify-end pt-3 text-sm font-medium">
-		<button
-			class="px-3.5 py-1.5 text-sm font-medium bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full"
-			type="submit"
-		>
-			{$i18n.t('Save')}
-		</button>
-	</div>
-</form>

src/lib/components/admin/Settings/WebSearch.svelte

@@ -23,7 +23,8 @@
 		'searchapi',
 		'duckduckgo',
 		'tavily',
-		'jina'
+		'jina',
+		'bing'
 	];
 
 	let youtubeLanguage = 'en';
@@ -234,6 +235,46 @@
 									bind:value={webConfig.search.tavily_api_key}
 								/>
 							</div>
+						{:else if webConfig.search.engine === 'jina'}
+							<div>
+								<div class=" self-center text-xs font-medium mb-1">
+									{$i18n.t('Jina API Key')}
+								</div>
+
+								<SensitiveInput
+									placeholder={$i18n.t('Enter Jina API Key')}
+									bind:value={webConfig.search.jina_api_key}
+								/>
+							</div>
+						{:else if webConfig.search.engine === 'bing'}
+							<div>
+								<div class=" self-center text-xs font-medium mb-1">
+									{$i18n.t('Bing Search V7 Endpoint')}
+								</div>
+
+								<div class="flex w-full">
+									<div class="flex-1">
+										<input
+											class="w-full rounded-lg py-2 px-4 text-sm bg-gray-50 dark:text-gray-300 dark:bg-gray-850 outline-none"
+											type="text"
+											placeholder={$i18n.t('Enter Bing Search V7 Endpoint')}
+											bind:value={webConfig.search.bing_search_v7_endpoint}
+											autocomplete="off"
+										/>
+									</div>
+								</div>
+							</div>
+
+							<div class="mt-2">
+								<div class=" self-center text-xs font-medium mb-1">
+									{$i18n.t('Bing Search V7 Subscription Key')}
+								</div>
+
+								<SensitiveInput
+									placeholder={$i18n.t('Enter Bing Search V7 Subscription Key')}
+									bind:value={webConfig.search.bing_search_v7_subscription_key}
+								/>
+							</div>
 						{/if}
 					</div>
 				{/if}
@@ -285,12 +326,12 @@
 						<button
 							class="p-1 px-3 text-xs flex rounded transition"
 							on:click={() => {
-								webConfig.ssl_verification = !webConfig.ssl_verification;
+								webConfig.web_loader_ssl_verification = !webConfig.web_loader_ssl_verification;
 								submitHandler();
 							}}
 							type="button"
 						>
-							{#if webConfig.ssl_verification === true}
+							{#if webConfig.web_loader_ssl_verification === false}
 								<span class="ml-2 self-center">{$i18n.t('On')}</span>
 							{:else}
 								<span class="ml-2 self-center">{$i18n.t('Off')}</span>

src/lib/components/admin/Users.svelte

@@ -0,0 +1,110 @@
+<script>
+	import { getContext, tick, onMount } from 'svelte';
+	import { toast } from 'svelte-sonner';
+
+	import { goto } from '$app/navigation';
+	import { user } from '$lib/stores';
+
+	import { getUsers } from '$lib/apis/users';
+
+	import UserList from './Users/UserList.svelte';
+	import Groups from './Users/Groups.svelte';
+
+	const i18n = getContext('i18n');
+
+	let users = [];
+
+	let selectedTab = 'overview';
+	let loaded = false;
+
+	$: if (selectedTab) {
+		getUsersHandler();
+	}
+
+	const getUsersHandler = async () => {
+		users = await getUsers(localStorage.token);
+	};
+
+	onMount(async () => {
+		if ($user?.role !== 'admin') {
+			await goto('/');
+		} else {
+			users = await getUsers(localStorage.token);
+		}
+		loaded = true;
+
+		const containerElement = document.getElementById('users-tabs-container');
+
+		if (containerElement) {
+			containerElement.addEventListener('wheel', function (event) {
+				if (event.deltaY !== 0) {
+					// Adjust horizontal scroll position based on vertical scroll
+					containerElement.scrollLeft += event.deltaY;
+				}
+			});
+		}
+	});
+</script>
+
+<div class="flex flex-col lg:flex-row w-full h-full pb-2 lg:space-x-4">
+	<div
+		id="users-tabs-container"
+		class=" flex flex-row overflow-x-auto gap-2.5 max-w-full lg:gap-1 lg:flex-col lg:flex-none lg:w-40 dark:text-gray-200 text-sm font-medium text-left scrollbar-none"
+	>
+		<button
+			class="px-0.5 py-1 min-w-fit rounded-lg lg:flex-none flex text-right transition {selectedTab ===
+			'overview'
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+			on:click={() => {
+				selectedTab = 'overview';
+			}}
+		>
+			<div class=" self-center mr-2">
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 16 16"
+					fill="currentColor"
+					class="size-4"
+				>
+					<path
+						d="M8.5 4.5a2.5 2.5 0 1 1-5 0 2.5 2.5 0 0 1 5 0ZM10.9 12.006c.11.542-.348.994-.9.994H2c-.553 0-1.01-.452-.902-.994a5.002 5.002 0 0 1 9.803 0ZM14.002 12h-1.59a2.556 2.556 0 0 0-.04-.29 6.476 6.476 0 0 0-1.167-2.603 3.002 3.002 0 0 1 3.633 1.911c.18.522-.283.982-.836.982ZM12 8a2 2 0 1 0 0-4 2 2 0 0 0 0 4Z"
+					/>
+				</svg>
+			</div>
+			<div class=" self-center">{$i18n.t('Overview')}</div>
+		</button>
+
+		<button
+			class="px-0.5 py-1 min-w-fit rounded-lg lg:flex-none flex text-right transition {selectedTab ===
+			'groups'
+				? ''
+				: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+			on:click={() => {
+				selectedTab = 'groups';
+			}}
+		>
+			<div class=" self-center mr-2">
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 16 16"
+					fill="currentColor"
+					class="size-4"
+				>
+					<path
+						d="M8 8a2.5 2.5 0 1 0 0-5 2.5 2.5 0 0 0 0 5ZM3.156 11.763c.16-.629.44-1.21.813-1.72a2.5 2.5 0 0 0-2.725 1.377c-.136.287.102.58.418.58h1.449c.01-.077.025-.156.045-.237ZM12.847 11.763c.02.08.036.16.046.237h1.446c.316 0 .554-.293.417-.579a2.5 2.5 0 0 0-2.722-1.378c.374.51.653 1.09.813 1.72ZM14 7.5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0ZM3.5 9a1.5 1.5 0 1 0 0-3 1.5 1.5 0 0 0 0 3ZM5 13c-.552 0-1.013-.455-.876-.99a4.002 4.002 0 0 1 7.753 0c.136.535-.324.99-.877.99H5Z"
+					/>
+				</svg>
+			</div>
+			<div class=" self-center">{$i18n.t('Groups')}</div>
+		</button>
+	</div>
+
+	<div class="flex-1 mt-1 lg:mt-0 overflow-y-scroll">
+		{#if selectedTab === 'overview'}
+			<UserList {users} />
+		{:else if selectedTab === 'groups'}
+			<Groups {users} />
+		{/if}
+	</div>
+</div>

src/lib/components/admin/Users/Groups.svelte

@@ -0,0 +1,237 @@
+<script>
+	import { toast } from 'svelte-sonner';
+	import dayjs from 'dayjs';
+	import relativeTime from 'dayjs/plugin/relativeTime';
+	dayjs.extend(relativeTime);
+
+	import { onMount, getContext } from 'svelte';
+	import { goto } from '$app/navigation';
+
+	import { WEBUI_NAME, config, user, showSidebar, knowledge } from '$lib/stores';
+	import { WEBUI_BASE_URL } from '$lib/constants';
+
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+	import Plus from '$lib/components/icons/Plus.svelte';
+	import Badge from '$lib/components/common/Badge.svelte';
+	import UsersSolid from '$lib/components/icons/UsersSolid.svelte';
+	import ChevronRight from '$lib/components/icons/ChevronRight.svelte';
+	import EllipsisHorizontal from '$lib/components/icons/EllipsisHorizontal.svelte';
+	import User from '$lib/components/icons/User.svelte';
+	import UserCircleSolid from '$lib/components/icons/UserCircleSolid.svelte';
+	import GroupModal from './Groups/EditGroupModal.svelte';
+	import Pencil from '$lib/components/icons/Pencil.svelte';
+	import GroupItem from './Groups/GroupItem.svelte';
+	import AddGroupModal from './Groups/AddGroupModal.svelte';
+	import { createNewGroup, getGroups } from '$lib/apis/groups';
+	import { getUserDefaultPermissions, updateUserDefaultPermissions } from '$lib/apis/users';
+
+	const i18n = getContext('i18n');
+
+	let loaded = false;
+
+	export let users = [];
+
+	let groups = [];
+	let filteredGroups;
+
+	$: filteredGroups = groups.filter((user) => {
+		if (search === '') {
+			return true;
+		} else {
+			let name = user.name.toLowerCase();
+			const query = search.toLowerCase();
+			return name.includes(query);
+		}
+	});
+
+	let search = '';
+	let defaultPermissions = {
+		workspace: {
+			models: false,
+			knowledge: false,
+			prompts: false,
+			tools: false
+		},
+		chat: {
+			file_upload: true,
+			delete: true,
+			edit: true,
+			temporary: true
+		}
+	};
+
+	let showCreateGroupModal = false;
+	let showDefaultPermissionsModal = false;
+
+	const setGroups = async () => {
+		groups = await getGroups(localStorage.token);
+	};
+
+	const addGroupHandler = async (group) => {
+		const res = await createNewGroup(localStorage.token, group).catch((error) => {
+			toast.error(error);
+			return null;
+		});
+
+		if (res) {
+			toast.success($i18n.t('Group created successfully'));
+			groups = await getGroups(localStorage.token);
+		}
+	};
+
+	const updateDefaultPermissionsHandler = async (group) => {
+		console.log(group.permissions);
+
+		const res = await updateUserDefaultPermissions(localStorage.token, group.permissions).catch(
+			(error) => {
+				toast.error(error);
+				return null;
+			}
+		);
+
+		if (res) {
+			toast.success($i18n.t('Default permissions updated successfully'));
+			defaultPermissions = await getUserDefaultPermissions(localStorage.token);
+		}
+	};
+
+	onMount(async () => {
+		if ($user?.role !== 'admin') {
+			await goto('/');
+		} else {
+			await setGroups();
+			defaultPermissions = await getUserDefaultPermissions(localStorage.token);
+		}
+		loaded = true;
+	});
+</script>
+
+{#if loaded}
+	<AddGroupModal bind:show={showCreateGroupModal} onSubmit={addGroupHandler} />
+	<div class="mt-0.5 mb-2 gap-1 flex flex-col md:flex-row justify-between">
+		<div class="flex md:self-center text-lg font-medium px-0.5">
+			{$i18n.t('Groups')}
+			<div class="flex self-center w-[1px] h-6 mx-2.5 bg-gray-50 dark:bg-gray-850" />
+
+			<span class="text-lg font-medium text-gray-500 dark:text-gray-300">{groups.length}</span>
+		</div>
+
+		<div class="flex gap-1">
+			<div class=" flex w-full space-x-2">
+				<div class="flex flex-1">
+					<div class=" self-center ml-1 mr-3">
+						<svg
+							xmlns="http://www.w3.org/2000/svg"
+							viewBox="0 0 20 20"
+							fill="currentColor"
+							class="w-4 h-4"
+						>
+							<path
+								fill-rule="evenodd"
+								d="M9 3.5a5.5 5.5 0 100 11 5.5 5.5 0 000-11zM2 9a7 7 0 1112.452 4.391l3.328 3.329a.75.75 0 11-1.06 1.06l-3.329-3.328A7 7 0 012 9z"
+								clip-rule="evenodd"
+							/>
+						</svg>
+					</div>
+					<input
+						class=" w-full text-sm pr-4 py-1 rounded-r-xl outline-none bg-transparent"
+						bind:value={search}
+						placeholder={$i18n.t('Search')}
+					/>
+				</div>
+
+				<div>
+					<Tooltip content={$i18n.t('Create Group')}>
+						<button
+							class=" p-2 rounded-xl hover:bg-gray-100 dark:bg-gray-900 dark:hover:bg-gray-850 transition font-medium text-sm flex items-center space-x-1"
+							on:click={() => {
+								showCreateGroupModal = !showCreateGroupModal;
+							}}
+						>
+							<Plus className="size-3.5" />
+						</button>
+					</Tooltip>
+				</div>
+			</div>
+		</div>
+	</div>
+
+	<div>
+		{#if filteredGroups.length === 0}
+			<div class="flex flex-col items-center justify-center h-40">
+				<div class=" text-xl font-medium">
+					{$i18n.t('Organize your users')}
+				</div>
+
+				<div class="mt-1 text-sm dark:text-gray-300">
+					{$i18n.t('Use groups to group your users and assign permissions.')}
+				</div>
+
+				<div class="mt-3">
+					<button
+						class=" px-4 py-1.5 text-sm rounded-full bg-black hover:bg-gray-800 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition font-medium flex items-center space-x-1"
+						aria-label={$i18n.t('Create Group')}
+						on:click={() => {
+							showCreateGroupModal = true;
+						}}
+					>
+						{$i18n.t('Create Group')}
+					</button>
+				</div>
+			</div>
+		{:else}
+			<div>
+				<div class=" flex items-center gap-3 justify-between text-xs uppercase px-1 font-bold">
+					<div class="w-full">Group</div>
+
+					<div class="w-full">Users</div>
+
+					<div class="w-full"></div>
+				</div>
+
+				<hr class="mt-1.5 border-gray-50 dark:border-gray-850" />
+
+				{#each filteredGroups as group}
+					<div class="my-2">
+						<GroupItem {group} {users} {setGroups} />
+					</div>
+				{/each}
+			</div>
+		{/if}
+
+		<hr class="mb-2 border-gray-50 dark:border-gray-850" />
+
+		<GroupModal
+			bind:show={showDefaultPermissionsModal}
+			tabs={['permissions']}
+			bind:permissions={defaultPermissions}
+			custom={false}
+			onSubmit={updateDefaultPermissionsHandler}
+		/>
+
+		<button
+			class="flex items-center justify-between rounded-lg w-full transition pt-1"
+			on:click={() => {
+				showDefaultPermissionsModal = true;
+			}}
+		>
+			<div class="flex items-center gap-2.5">
+				<div class="p-1.5 bg-black/5 dark:bg-white/10 rounded-full">
+					<UsersSolid className="size-4" />
+				</div>
+
+				<div class="text-left">
+					<div class=" text-sm font-medium">{$i18n.t('Default permissions')}</div>
+
+					<div class="flex text-xs mt-0.5">
+						{$i18n.t('applies to all users with the "user" role')}
+					</div>
+				</div>
+			</div>
+
+			<div>
+				<ChevronRight strokeWidth="2.5" />
+			</div>
+		</button>
+	</div>
+{/if}

src/lib/components/admin/Users/Groups/AddGroupModal.svelte

@@ -0,0 +1,149 @@
+<script lang="ts">
+	import { toast } from 'svelte-sonner';
+	import { getContext, onMount } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import Modal from '$lib/components/common/Modal.svelte';
+	import Textarea from '$lib/components/common/Textarea.svelte';
+	export let onSubmit: Function = () => {};
+	export let show = false;
+
+	let name = '';
+	let description = '';
+	let userIds = [];
+
+	let loading = false;
+
+	const submitHandler = async () => {
+		loading = true;
+
+		const group = {
+			name,
+			description
+		};
+
+		await onSubmit(group);
+
+		loading = false;
+		show = false;
+
+		name = '';
+		description = '';
+		userIds = [];
+	};
+
+	onMount(() => {
+		console.log('mounted');
+	});
+</script>
+
+<Modal size="sm" bind:show>
+	<div>
+		<div class=" flex justify-between dark:text-gray-100 px-5 pt-4 mb-1.5">
+			<div class=" text-lg font-medium self-center font-primary">
+				{$i18n.t('Add User Group')}
+			</div>
+			<button
+				class="self-center"
+				on:click={() => {
+					show = false;
+				}}
+			>
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 20 20"
+					fill="currentColor"
+					class="w-5 h-5"
+				>
+					<path
+						d="M6.28 5.22a.75.75 0 00-1.06 1.06L8.94 10l-3.72 3.72a.75.75 0 101.06 1.06L10 11.06l3.72 3.72a.75.75 0 101.06-1.06L11.06 10l3.72-3.72a.75.75 0 00-1.06-1.06L10 8.94 6.28 5.22z"
+					/>
+				</svg>
+			</button>
+		</div>
+
+		<div class="flex flex-col md:flex-row w-full px-4 pb-4 md:space-x-4 dark:text-gray-200">
+			<div class=" flex flex-col w-full sm:flex-row sm:justify-center sm:space-x-6">
+				<form
+					class="flex flex-col w-full"
+					on:submit={(e) => {
+						e.preventDefault();
+						submitHandler();
+					}}
+				>
+					<div class="px-1 flex flex-col w-full">
+						<div class="flex gap-2">
+							<div class="flex flex-col w-full">
+								<div class=" mb-0.5 text-xs text-gray-500">{$i18n.t('Name')}</div>
+
+								<div class="flex-1">
+									<input
+										class="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+										type="text"
+										bind:value={name}
+										placeholder={$i18n.t('Group Name')}
+										autocomplete="off"
+										required
+									/>
+								</div>
+							</div>
+						</div>
+
+						<div class="flex flex-col w-full mt-2">
+							<div class=" mb-1 text-xs text-gray-500">{$i18n.t('Description')}</div>
+
+							<div class="flex-1">
+								<Textarea
+									className="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none resize-none"
+									rows={2}
+									bind:value={description}
+									placeholder={$i18n.t('Group Description')}
+								/>
+							</div>
+						</div>
+					</div>
+
+					<div class="flex justify-end pt-3 text-sm font-medium gap-1.5">
+						<button
+							class="px-3.5 py-1.5 text-sm font-medium bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full flex flex-row space-x-1 items-center {loading
+								? ' cursor-not-allowed'
+								: ''}"
+							type="submit"
+							disabled={loading}
+						>
+							{$i18n.t('Create')}
+
+							{#if loading}
+								<div class="ml-2 self-center">
+									<svg
+										class=" w-4 h-4"
+										viewBox="0 0 24 24"
+										fill="currentColor"
+										xmlns="http://www.w3.org/2000/svg"
+										><style>
+											.spinner_ajPY {
+												transform-origin: center;
+												animation: spinner_AtaB 0.75s infinite linear;
+											}
+											@keyframes spinner_AtaB {
+												100% {
+													transform: rotate(360deg);
+												}
+											}
+										</style><path
+											d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
+											opacity=".25"
+										/><path
+											d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
+											class="spinner_ajPY"
+										/></svg
+									>
+								</div>
+							{/if}
+						</button>
+					</div>
+				</form>
+			</div>
+		</div>
+	</div>
+</Modal>

src/lib/components/admin/Users/Groups/Display.svelte

@@ -0,0 +1,61 @@
+<script lang="ts">
+	import { getContext } from 'svelte';
+	import Textarea from '$lib/components/common/Textarea.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+
+	const i18n = getContext('i18n');
+
+	export let name = '';
+	export let color = '';
+	export let description = '';
+</script>
+
+<div class="flex gap-2">
+	<div class="flex flex-col w-full">
+		<div class=" mb-0.5 text-xs text-gray-500">{$i18n.t('Name')}</div>
+
+		<div class="flex-1">
+			<input
+				class="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+				type="text"
+				bind:value={name}
+				placeholder={$i18n.t('Group Name')}
+				autocomplete="off"
+				required
+			/>
+		</div>
+	</div>
+</div>
+
+<!-- <div class="flex flex-col w-full mt-2">
+	<div class=" mb-1 text-xs text-gray-500">{$i18n.t('Color')}</div>
+
+	<div class="flex-1">
+		<Tooltip content={$i18n.t('Hex Color - Leave empty for default color')} placement="top-start">
+			<div class="flex gap-0.5">
+				<div class="text-gray-500">#</div>
+
+				<input
+					class="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+					type="text"
+					bind:value={color}
+					placeholder={$i18n.t('Hex Color')}
+					autocomplete="off"
+				/>
+			</div>
+		</Tooltip>
+	</div>
+</div> -->
+
+<div class="flex flex-col w-full mt-2">
+	<div class=" mb-1 text-xs text-gray-500">{$i18n.t('Description')}</div>
+
+	<div class="flex-1">
+		<Textarea
+			className="w-full text-sm bg-transparent placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none resize-none"
+			rows={4}
+			bind:value={description}
+			placeholder={$i18n.t('Group Description')}
+		/>
+	</div>
+</div>

src/lib/components/admin/Users/Groups/EditGroupModal.svelte

@@ -0,0 +1,328 @@
+<script lang="ts">
+	import { toast } from 'svelte-sonner';
+	import { getContext, onMount } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import Modal from '$lib/components/common/Modal.svelte';
+	import Display from './Display.svelte';
+	import Permissions from './Permissions.svelte';
+	import Users from './Users.svelte';
+	import UserPlusSolid from '$lib/components/icons/UserPlusSolid.svelte';
+	import WrenchSolid from '$lib/components/icons/WrenchSolid.svelte';
+
+	export let onSubmit: Function = () => {};
+	export let onDelete: Function = () => {};
+
+	export let show = false;
+	export let edit = false;
+
+	export let users = [];
+	export let group = null;
+
+	export let custom = true;
+
+	export let tabs = ['general', 'permissions', 'users'];
+
+	let selectedTab = 'general';
+	let loading = false;
+
+	export let name = '';
+	export let description = '';
+
+	export let permissions = {
+		workspace: {
+			models: false,
+			knowledge: false,
+			prompts: false,
+			tools: false
+		},
+		chat: {
+			file_upload: true,
+			delete: true,
+			edit: true,
+			temporary: true
+		}
+	};
+	export let userIds = [];
+
+	const submitHandler = async () => {
+		loading = true;
+
+		const group = {
+			name,
+			description,
+			permissions,
+			user_ids: userIds
+		};
+
+		await onSubmit(group);
+
+		loading = false;
+		show = false;
+	};
+
+	const init = () => {
+		if (group) {
+			name = group.name;
+			description = group.description;
+			permissions = group?.permissions ?? {
+				workspace: {
+					models: false,
+					knowledge: false,
+					prompts: false,
+					tools: false
+				},
+				chat: {
+					file_upload: true,
+					delete: true,
+					edit: true,
+					temporary: true
+				}
+			};
+			userIds = group?.user_ids ?? [];
+		}
+	};
+
+	$: if (show) {
+		init();
+	}
+
+	onMount(() => {
+		console.log(tabs);
+		selectedTab = tabs[0];
+		init();
+	});
+</script>
+
+<Modal size="md" bind:show>
+	<div>
+		<div class=" flex justify-between dark:text-gray-100 px-5 pt-4 mb-1.5">
+			<div class=" text-lg font-medium self-center font-primary">
+				{#if custom}
+					{#if edit}
+						{$i18n.t('Edit User Group')}
+					{:else}
+						{$i18n.t('Add User Group')}
+					{/if}
+				{:else}
+					{$i18n.t('Edit Default Permissions')}
+				{/if}
+			</div>
+			<button
+				class="self-center"
+				on:click={() => {
+					show = false;
+				}}
+			>
+				<svg
+					xmlns="http://www.w3.org/2000/svg"
+					viewBox="0 0 20 20"
+					fill="currentColor"
+					class="w-5 h-5"
+				>
+					<path
+						d="M6.28 5.22a.75.75 0 00-1.06 1.06L8.94 10l-3.72 3.72a.75.75 0 101.06 1.06L10 11.06l3.72 3.72a.75.75 0 101.06-1.06L11.06 10l3.72-3.72a.75.75 0 00-1.06-1.06L10 8.94 6.28 5.22z"
+					/>
+				</svg>
+			</button>
+		</div>
+
+		<div class="flex flex-col md:flex-row w-full px-4 pb-4 md:space-x-4 dark:text-gray-200">
+			<div class=" flex flex-col w-full sm:flex-row sm:justify-center sm:space-x-6">
+				<form
+					class="flex flex-col w-full"
+					on:submit={(e) => {
+						e.preventDefault();
+						submitHandler();
+					}}
+				>
+					<div class="flex flex-col lg:flex-row w-full h-full pb-2 lg:space-x-4">
+						<div
+							id="admin-settings-tabs-container"
+							class="tabs flex flex-row overflow-x-auto gap-2.5 max-w-full lg:gap-1 lg:flex-col lg:flex-none lg:w-40 dark:text-gray-200 text-sm font-medium text-left scrollbar-none"
+						>
+							{#if tabs.includes('general')}
+								<button
+									class="px-0.5 py-1 max-w-fit w-fit rounded-lg flex-1 lg:flex-none flex text-right transition {selectedTab ===
+									'general'
+										? ''
+										: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+									on:click={() => {
+										selectedTab = 'general';
+									}}
+									type="button"
+								>
+									<div class=" self-center mr-2">
+										<svg
+											xmlns="http://www.w3.org/2000/svg"
+											viewBox="0 0 16 16"
+											fill="currentColor"
+											class="w-4 h-4"
+										>
+											<path
+												fill-rule="evenodd"
+												d="M6.955 1.45A.5.5 0 0 1 7.452 1h1.096a.5.5 0 0 1 .497.45l.17 1.699c.484.12.94.312 1.356.562l1.321-1.081a.5.5 0 0 1 .67.033l.774.775a.5.5 0 0 1 .034.67l-1.08 1.32c.25.417.44.873.561 1.357l1.699.17a.5.5 0 0 1 .45.497v1.096a.5.5 0 0 1-.45.497l-1.699.17c-.12.484-.312.94-.562 1.356l1.082 1.322a.5.5 0 0 1-.034.67l-.774.774a.5.5 0 0 1-.67.033l-1.322-1.08c-.416.25-.872.44-1.356.561l-.17 1.699a.5.5 0 0 1-.497.45H7.452a.5.5 0 0 1-.497-.45l-.17-1.699a4.973 4.973 0 0 1-1.356-.562L4.108 13.37a.5.5 0 0 1-.67-.033l-.774-.775a.5.5 0 0 1-.034-.67l1.08-1.32a4.971 4.971 0 0 1-.561-1.357l-1.699-.17A.5.5 0 0 1 1 8.548V7.452a.5.5 0 0 1 .45-.497l1.699-.17c.12-.484.312-.94.562-1.356L2.629 4.107a.5.5 0 0 1 .034-.67l.774-.774a.5.5 0 0 1 .67-.033L5.43 3.71a4.97 4.97 0 0 1 1.356-.561l.17-1.699ZM6 8c0 .538.212 1.026.558 1.385l.057.057a2 2 0 0 0 2.828-2.828l-.058-.056A2 2 0 0 0 6 8Z"
+												clip-rule="evenodd"
+											/>
+										</svg>
+									</div>
+									<div class=" self-center">{$i18n.t('General')}</div>
+								</button>
+							{/if}
+
+							{#if tabs.includes('permissions')}
+								<button
+									class="px-0.5 py-1 max-w-fit w-fit rounded-lg flex-1 lg:flex-none flex text-right transition {selectedTab ===
+									'permissions'
+										? ''
+										: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+									on:click={() => {
+										selectedTab = 'permissions';
+									}}
+									type="button"
+								>
+									<div class=" self-center mr-2">
+										<WrenchSolid />
+									</div>
+									<div class=" self-center">{$i18n.t('Permissions')}</div>
+								</button>
+							{/if}
+
+							{#if tabs.includes('users')}
+								<button
+									class="px-0.5 py-1 max-w-fit w-fit rounded-lg flex-1 lg:flex-none flex text-right transition {selectedTab ===
+									'users'
+										? ''
+										: ' text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+									on:click={() => {
+										selectedTab = 'users';
+									}}
+									type="button"
+								>
+									<div class=" self-center mr-2">
+										<UserPlusSolid />
+									</div>
+									<div class=" self-center">{$i18n.t('Users')} ({userIds.length})</div>
+								</button>
+							{/if}
+						</div>
+
+						<div
+							class="flex-1 mt-1 lg:mt-1 lg:h-[22rem] lg:max-h-[22rem] overflow-y-auto scrollbar-hidden"
+						>
+							{#if selectedTab == 'general'}
+								<Display bind:name bind:description />
+							{:else if selectedTab == 'permissions'}
+								<Permissions bind:permissions />
+							{:else if selectedTab == 'users'}
+								<Users bind:userIds {users} />
+							{/if}
+						</div>
+					</div>
+
+					<!-- <div
+						class=" tabs flex flex-row overflow-x-auto gap-2.5 text-sm font-medium border-b border-b-gray-800 scrollbar-hidden"
+					>
+						{#if tabs.includes('display')}
+							<button
+								class="px-0.5 pb-1.5 min-w-fit flex text-right transition border-b-2 {selectedTab ===
+								'display'
+									? ' dark:border-white'
+									: 'border-transparent text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+								on:click={() => {
+									selectedTab = 'display';
+								}}
+								type="button"
+							>
+								{$i18n.t('Display')}
+							</button>
+						{/if}
+
+						{#if tabs.includes('permissions')}
+							<button
+								class="px-0.5 pb-1.5 min-w-fit flex text-right transition border-b-2 {selectedTab ===
+								'permissions'
+									? '  dark:border-white'
+									: 'border-transparent text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+								on:click={() => {
+									selectedTab = 'permissions';
+								}}
+								type="button"
+							>
+								{$i18n.t('Permissions')}
+							</button>
+						{/if}
+
+						{#if tabs.includes('users')}
+							<button
+								class="px-0.5 pb-1.5 min-w-fit flex text-right transition border-b-2 {selectedTab ===
+								'users'
+									? ' dark:border-white'
+									: ' border-transparent text-gray-300 dark:text-gray-600 hover:text-gray-700 dark:hover:text-white'}"
+								on:click={() => {
+									selectedTab = 'users';
+								}}
+								type="button"
+							>
+								{$i18n.t('Users')} ({userIds.length})
+							</button>
+						{/if}
+					</div> -->
+
+					<div class="flex justify-end pt-3 text-sm font-medium gap-1.5">
+						{#if edit}
+							<button
+								class="px-3.5 py-1.5 text-sm font-medium dark:bg-black dark:hover:bg-gray-900 dark:text-white bg-white text-black hover:bg-gray-100 transition rounded-full flex flex-row space-x-1 items-center"
+								type="button"
+								on:click={() => {
+									onDelete();
+									show = false;
+								}}
+							>
+								{$i18n.t('Delete')}
+							</button>
+						{/if}
+
+						<button
+							class="px-3.5 py-1.5 text-sm font-medium bg-black hover:bg-gray-900 text-white dark:bg-white dark:text-black dark:hover:bg-gray-100 transition rounded-full flex flex-row space-x-1 items-center {loading
+								? ' cursor-not-allowed'
+								: ''}"
+							type="submit"
+							disabled={loading}
+						>
+							{$i18n.t('Save')}
+
+							{#if loading}
+								<div class="ml-2 self-center">
+									<svg
+										class=" w-4 h-4"
+										viewBox="0 0 24 24"
+										fill="currentColor"
+										xmlns="http://www.w3.org/2000/svg"
+										><style>
+											.spinner_ajPY {
+												transform-origin: center;
+												animation: spinner_AtaB 0.75s infinite linear;
+											}
+											@keyframes spinner_AtaB {
+												100% {
+													transform: rotate(360deg);
+												}
+											}
+										</style><path
+											d="M12,1A11,11,0,1,0,23,12,11,11,0,0,0,12,1Zm0,19a8,8,0,1,1,8-8A8,8,0,0,1,12,20Z"
+											opacity=".25"
+										/><path
+											d="M10.14,1.16a11,11,0,0,0-9,8.92A1.59,1.59,0,0,0,2.46,12,1.52,1.52,0,0,0,4.11,10.7a8,8,0,0,1,6.66-6.61A1.42,1.42,0,0,0,12,2.69h0A1.57,1.57,0,0,0,10.14,1.16Z"
+											class="spinner_ajPY"
+										/></svg
+									>
+								</div>
+							{/if}
+						</button>
+					</div>
+				</form>
+			</div>
+		</div>
+	</div>
+</Modal>

src/lib/components/admin/Users/Groups/GroupItem.svelte

@@ -0,0 +1,84 @@
+<script>
+	import { toast } from 'svelte-sonner';
+	import { getContext } from 'svelte';
+
+	const i18n = getContext('i18n');
+
+	import { deleteGroupById, updateGroupById } from '$lib/apis/groups';
+
+	import Pencil from '$lib/components/icons/Pencil.svelte';
+	import User from '$lib/components/icons/User.svelte';
+	import UserCircleSolid from '$lib/components/icons/UserCircleSolid.svelte';
+	import GroupModal from './EditGroupModal.svelte';
+
+	export let users = [];
+	export let group = {
+		name: 'Admins',
+		user_ids: [1, 2, 3]
+	};
+
+	export let setGroups = () => {};
+
+	let showEdit = false;
+
+	const updateHandler = async (_group) => {
+		const res = await updateGroupById(localStorage.token, group.id, _group).catch((error) => {
+			toast.error(error);
+			return null;
+		});
+
+		if (res) {
+			toast.success($i18n.t('Group updated successfully'));
+			setGroups();
+		}
+	};
+
+	const deleteHandler = async () => {
+		const res = await deleteGroupById(localStorage.token, group.id).catch((error) => {
+			toast.error(error);
+			return null;
+		});
+
+		if (res) {
+			toast.success($i18n.t('Group deleted successfully'));
+			setGroups();
+		}
+	};
+</script>
+
+<GroupModal
+	bind:show={showEdit}
+	edit
+	{users}
+	{group}
+	onSubmit={updateHandler}
+	onDelete={deleteHandler}
+/>
+
+<button
+	class="flex items-center gap-3 justify-between px-1 text-xs w-full transition"
+	on:click={() => {
+		showEdit = true;
+	}}
+>
+	<div class="flex items-center gap-1.5 w-full font-medium">
+		<div>
+			<UserCircleSolid className="size-4" />
+		</div>
+		{group.name}
+	</div>
+
+	<div class="flex items-center gap-1.5 w-full font-medium">
+		{group.user_ids.length}
+
+		<div>
+			<User className="size-3.5" />
+		</div>
+	</div>
+
+	<div class="w-full flex justify-end">
+		<div class=" rounded-lg p-1 hover:bg-gray-100 dark:hover:bg-gray-850 transition">
+			<Pencil className="size-3.5" />
+		</div>
+	</div>
+</button>

src/lib/components/admin/Users/Groups/Permissions.svelte

@@ -0,0 +1,204 @@
+<script lang="ts">
+	import { getContext } from 'svelte';
+	const i18n = getContext('i18n');
+
+	import Switch from '$lib/components/common/Switch.svelte';
+	import Tooltip from '$lib/components/common/Tooltip.svelte';
+
+	export let permissions = {
+		workspace: {
+			models: false,
+			knowledge: false,
+			prompts: false,
+			tools: false
+		},
+		chat: {
+			delete: true,
+			edit: true,
+			temporary: true,
+			file_upload: true
+		}
+	};
+</script>
+
+<div>
+	<!-- <div>
+		<div class=" mb-2 text-sm font-medium">{$i18n.t('Model Permissions')}</div>
+
+		<div class="mb-2">
+			<div class="flex justify-between items-center text-xs pr-2">
+				<div class=" text-xs font-medium">{$i18n.t('Model Filtering')}</div>
+
+				<Switch bind:state={permissions.model.filter} />
+			</div>
+		</div>
+
+		{#if permissions.model.filter}
+			<div class="mb-2">
+				<div class=" space-y-1.5">
+					<div class="flex flex-col w-full">
+						<div class="mb-1 flex justify-between">
+							<div class="text-xs text-gray-500">{$i18n.t('Model IDs')}</div>
+						</div>
+
+						{#if model_ids.length > 0}
+							<div class="flex flex-col">
+								{#each model_ids as modelId, modelIdx}
+									<div class=" flex gap-2 w-full justify-between items-center">
+										<div class=" text-sm flex-1 rounded-lg">
+											{modelId}
+										</div>
+										<div class="flex-shrink-0">
+											<button
+												type="button"
+												on:click={() => {
+													model_ids = model_ids.filter((_, idx) => idx !== modelIdx);
+												}}
+											>
+												<Minus strokeWidth="2" className="size-3.5" />
+											</button>
+										</div>
+									</div>
+								{/each}
+							</div>
+						{:else}
+							<div class="text-gray-500 text-xs text-center py-2 px-10">
+								{$i18n.t('No model IDs')}
+							</div>
+						{/if}
+					</div>
+				</div>
+				<hr class=" border-gray-100 dark:border-gray-700/10 mt-2.5 mb-1 w-full" />
+
+				<div class="flex items-center">
+					<select
+						class="w-full py-1 text-sm rounded-lg bg-transparent {selectedModelId
+							? ''
+							: 'text-gray-500'} placeholder:text-gray-300 dark:placeholder:text-gray-700 outline-none"
+						bind:value={selectedModelId}
+					>
+						<option value="">{$i18n.t('Select a model')}</option>
+						{#each $models.filter((m) => m?.owned_by !== 'arena') as model}
+							<option value={model.id} class="bg-gray-50 dark:bg-gray-700">{model.name}</option>
+						{/each}
+					</select>
+
+					<div>
+						<button
+							type="button"
+							on:click={() => {
+								if (selectedModelId && !permissions.model.model_ids.includes(selectedModelId)) {
+									permissions.model.model_ids = [...permissions.model.model_ids, selectedModelId];
+									selectedModelId = '';
+								}
+							}}
+						>
+							<Plus className="size-3.5" strokeWidth="2" />
+						</button>
+					</div>
+				</div>
+			</div>
+		{/if}
+
+		<div class=" space-y-1 mb-3">
+			<div class="">
+				<div class="flex justify-between items-center text-xs">
+					<div class=" text-xs font-medium">{$i18n.t('Default Model')}</div>
+				</div>
+			</div>
+
+			<div class="flex-1 mr-2">
+				<select
+					class="w-full bg-transparent outline-none py-0.5 text-sm"
+					bind:value={permissions.model.default_id}
+					placeholder="Select a model"
+				>
+					<option value="" disabled selected>{$i18n.t('Select a model')}</option>
+					{#each permissions.model.filter ? $models.filter( (model) => filterModelIds.includes(model.id) ) : $models.filter((model) => model.id) as model}
+						<option value={model.id} class="bg-gray-100 dark:bg-gray-700">{model.name}</option>
+					{/each}
+				</select>
+			</div>
+		</div>
+	</div>
+
+	<hr class=" border-gray-50 dark:border-gray-850 my-2" /> -->
+
+	<div>
+		<div class=" mb-2 text-sm font-medium">{$i18n.t('Workspace Permissions')}</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Models Access')}
+			</div>
+			<Switch bind:state={permissions.workspace.models} />
+		</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Knowledge Access')}
+			</div>
+			<Switch bind:state={permissions.workspace.knowledge} />
+		</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Prompts Access')}
+			</div>
+			<Switch bind:state={permissions.workspace.prompts} />
+		</div>
+
+		<div class=" ">
+			<Tooltip
+				className=" flex w-full justify-between my-2 pr-2"
+				content={$i18n.t(
+					'Warning: Enabling this will allow users to upload arbitrary code on the server.'
+				)}
+				placement="top-start"
+			>
+				<div class=" self-center text-xs font-medium">
+					{$i18n.t('Tools Access')}
+				</div>
+				<Switch bind:state={permissions.workspace.tools} />
+			</Tooltip>
+		</div>
+	</div>
+
+	<hr class=" border-gray-50 dark:border-gray-850 my-2" />
+
+	<div>
+		<div class=" mb-2 text-sm font-medium">{$i18n.t('Chat Permissions')}</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Allow File Upload')}
+			</div>
+
+			<Switch bind:state={permissions.chat.file_upload} />
+		</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Allow Chat Delete')}
+			</div>
+
+			<Switch bind:state={permissions.chat.delete} />
+		</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Allow Chat Edit')}
+			</div>
+
+			<Switch bind:state={permissions.chat.edit} />
+		</div>
+
+		<div class="  flex w-full justify-between my-2 pr-2">
+			<div class=" self-center text-xs font-medium">
+				{$i18n.t('Allow Temporary Chat')}
+			</div>
+
+			<Switch bind:state={permissions.chat.temporary} />
+		</div>
+	</div>
+</div>