Merge branch 'dev' of github.com:open-webui/open-webui

CHANGELOG.md

@@ -5,6 +5,9 @@ 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).
 
+### Added
+- **🌐 Enhanced Translations**: Added Slovak language, improved Czech language.
+
 ## [0.4.8] - 2024-12-07
 
 ### Added

backend/open_webui/apps/audio/main.py

@@ -1,703 +0,0 @@
-import hashlib
-import json
-import logging
-import os
-import uuid
-from functools import lru_cache
-from pathlib import Path
-from pydub import AudioSegment
-from pydub.silence import split_on_silence
-
-import aiohttp
-import aiofiles
-import requests
-from open_webui.config import (
-    AUDIO_STT_ENGINE,
-    AUDIO_STT_MODEL,
-    AUDIO_STT_OPENAI_API_BASE_URL,
-    AUDIO_STT_OPENAI_API_KEY,
-    AUDIO_TTS_API_KEY,
-    AUDIO_TTS_ENGINE,
-    AUDIO_TTS_MODEL,
-    AUDIO_TTS_OPENAI_API_BASE_URL,
-    AUDIO_TTS_OPENAI_API_KEY,
-    AUDIO_TTS_SPLIT_ON,
-    AUDIO_TTS_VOICE,
-    AUDIO_TTS_AZURE_SPEECH_REGION,
-    AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT,
-    CACHE_DIR,
-    CORS_ALLOW_ORIGIN,
-    WHISPER_MODEL,
-    WHISPER_MODEL_AUTO_UPDATE,
-    WHISPER_MODEL_DIR,
-    AppConfig,
-)
-
-from open_webui.constants import ERROR_MESSAGES
-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
-from pydantic import BaseModel
-from open_webui.utils.utils import get_admin_user, get_verified_user
-
-# Constants
-MAX_FILE_SIZE_MB = 25
-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(
-    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,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-app.state.config = AppConfig()
-
-app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL
-app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY
-app.state.config.STT_ENGINE = AUDIO_STT_ENGINE
-app.state.config.STT_MODEL = AUDIO_STT_MODEL
-
-app.state.config.WHISPER_MODEL = WHISPER_MODEL
-app.state.faster_whisper_model = None
-
-app.state.config.TTS_OPENAI_API_BASE_URL = AUDIO_TTS_OPENAI_API_BASE_URL
-app.state.config.TTS_OPENAI_API_KEY = AUDIO_TTS_OPENAI_API_KEY
-app.state.config.TTS_ENGINE = AUDIO_TTS_ENGINE
-app.state.config.TTS_MODEL = AUDIO_TTS_MODEL
-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
-
-# setting device type for whisper model
-whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu"
-log.info(f"whisper_device_type: {whisper_device_type}")
-
-SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
-SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
-
-
-def set_faster_whisper_model(model: str, auto_update: bool = False):
-    if model and app.state.config.STT_ENGINE == "":
-        from faster_whisper import WhisperModel
-
-        faster_whisper_kwargs = {
-            "model_size_or_path": model,
-            "device": whisper_device_type,
-            "compute_type": "int8",
-            "download_root": WHISPER_MODEL_DIR,
-            "local_files_only": not auto_update,
-        }
-
-        try:
-            app.state.faster_whisper_model = WhisperModel(**faster_whisper_kwargs)
-        except Exception:
-            log.warning(
-                "WhisperModel initialization failed, attempting download with local_files_only=False"
-            )
-            faster_whisper_kwargs["local_files_only"] = False
-            app.state.faster_whisper_model = WhisperModel(**faster_whisper_kwargs)
-
-    else:
-        app.state.faster_whisper_model = None
-
-
-class TTSConfigForm(BaseModel):
-    OPENAI_API_BASE_URL: str
-    OPENAI_API_KEY: str
-    API_KEY: str
-    ENGINE: str
-    MODEL: str
-    VOICE: str
-    SPLIT_ON: str
-    AZURE_SPEECH_REGION: str
-    AZURE_SPEECH_OUTPUT_FORMAT: str
-
-
-class STTConfigForm(BaseModel):
-    OPENAI_API_BASE_URL: str
-    OPENAI_API_KEY: str
-    ENGINE: str
-    MODEL: str
-    WHISPER_MODEL: str
-
-
-class AudioConfigUpdateForm(BaseModel):
-    tts: TTSConfigForm
-    stt: STTConfigForm
-
-
-from pydub import AudioSegment
-from pydub.utils import mediainfo
-
-
-def is_mp4_audio(file_path):
-    """Check if the given file is an MP4 audio file."""
-    if not os.path.isfile(file_path):
-        print(f"File not found: {file_path}")
-        return False
-
-    info = mediainfo(file_path)
-    if (
-        info.get("codec_name") == "aac"
-        and info.get("codec_type") == "audio"
-        and info.get("codec_tag_string") == "mp4a"
-    ):
-        return True
-    return False
-
-
-def convert_mp4_to_wav(file_path, output_path):
-    """Convert MP4 audio file to WAV format."""
-    audio = AudioSegment.from_file(file_path, format="mp4")
-    audio.export(output_path, format="wav")
-    print(f"Converted {file_path} to {output_path}")
-
-
-@app.get("/config")
-async def get_audio_config(user=Depends(get_admin_user)):
-    return {
-        "tts": {
-            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
-            "API_KEY": app.state.config.TTS_API_KEY,
-            "ENGINE": app.state.config.TTS_ENGINE,
-            "MODEL": app.state.config.TTS_MODEL,
-            "VOICE": app.state.config.TTS_VOICE,
-            "SPLIT_ON": app.state.config.TTS_SPLIT_ON,
-            "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION,
-            "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
-        },
-        "stt": {
-            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
-            "ENGINE": app.state.config.STT_ENGINE,
-            "MODEL": app.state.config.STT_MODEL,
-            "WHISPER_MODEL": app.state.config.WHISPER_MODEL,
-        },
-    }
-
-
-@app.post("/config/update")
-async def update_audio_config(
-    form_data: AudioConfigUpdateForm, user=Depends(get_admin_user)
-):
-    app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL
-    app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY
-    app.state.config.TTS_API_KEY = form_data.tts.API_KEY
-    app.state.config.TTS_ENGINE = form_data.tts.ENGINE
-    app.state.config.TTS_MODEL = form_data.tts.MODEL
-    app.state.config.TTS_VOICE = form_data.tts.VOICE
-    app.state.config.TTS_SPLIT_ON = form_data.tts.SPLIT_ON
-    app.state.config.TTS_AZURE_SPEECH_REGION = form_data.tts.AZURE_SPEECH_REGION
-    app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = (
-        form_data.tts.AZURE_SPEECH_OUTPUT_FORMAT
-    )
-
-    app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL
-    app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY
-    app.state.config.STT_ENGINE = form_data.stt.ENGINE
-    app.state.config.STT_MODEL = form_data.stt.MODEL
-    app.state.config.WHISPER_MODEL = form_data.stt.WHISPER_MODEL
-    set_faster_whisper_model(form_data.stt.WHISPER_MODEL, WHISPER_MODEL_AUTO_UPDATE)
-
-    return {
-        "tts": {
-            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
-            "API_KEY": app.state.config.TTS_API_KEY,
-            "ENGINE": app.state.config.TTS_ENGINE,
-            "MODEL": app.state.config.TTS_MODEL,
-            "VOICE": app.state.config.TTS_VOICE,
-            "SPLIT_ON": app.state.config.TTS_SPLIT_ON,
-            "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION,
-            "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
-        },
-        "stt": {
-            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
-            "ENGINE": app.state.config.STT_ENGINE,
-            "MODEL": app.state.config.STT_MODEL,
-            "WHISPER_MODEL": app.state.config.WHISPER_MODEL,
-        },
-    }
-
-
-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()
-    name = hashlib.sha256(body).hexdigest()
-
-    file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
-    file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
-
-    # Check if the file already exists in the cache
-    if file_path.is_file():
-        return FileResponse(file_path)
-
-    if app.state.config.TTS_ENGINE == "openai":
-        headers = {}
-        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)
-            body["model"] = app.state.config.TTS_MODEL
-            body = json.dumps(body).encode("utf-8")
-        except Exception:
-            pass
-
-        try:
-            async with aiohttp.ClientSession() as session:
-                async with session.post(
-                    url=f"{app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech",
-                    data=body,
-                    headers=headers,
-                ) as r:
-                    r.raise_for_status()
-                    async with aiofiles.open(file_path, "wb") as f:
-                        await f.write(await r.read())
-
-                    async with aiofiles.open(file_body_path, "w") as f:
-                        await f.write(json.dumps(json.loads(body.decode("utf-8"))))
-
-            return FileResponse(file_path)
-
-        except Exception as e:
-            log.exception(e)
-            error_detail = "Open WebUI: Server Connection Error"
-            try:
-                if r.status != 200:
-                    res = await r.json()
-                    if "error" in res:
-                        error_detail = f"External: {res['error']['message']}"
-            except Exception:
-                error_detail = f"External: {e}"
-
-            raise HTTPException(
-                status_code=getattr(r, "status", 500),
-                detail=error_detail,
-            )
-
-    elif app.state.config.TTS_ENGINE == "elevenlabs":
-        try:
-            payload = json.loads(body.decode("utf-8"))
-        except Exception as e:
-            log.exception(e)
-            raise HTTPException(status_code=400, detail="Invalid JSON payload")
-
-        voice_id = payload.get("voice", "")
-        if voice_id not in get_available_voices():
-            raise HTTPException(
-                status_code=400,
-                detail="Invalid voice id",
-            )
-
-        url = f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}"
-        headers = {
-            "Accept": "audio/mpeg",
-            "Content-Type": "application/json",
-            "xi-api-key": app.state.config.TTS_API_KEY,
-        }
-        data = {
-            "text": payload["input"],
-            "model_id": app.state.config.TTS_MODEL,
-            "voice_settings": {"stability": 0.5, "similarity_boost": 0.5},
-        }
-
-        try:
-            async with aiohttp.ClientSession() as session:
-                async with session.post(url, json=data, headers=headers) as r:
-                    r.raise_for_status()
-                    async with aiofiles.open(file_path, "wb") as f:
-                        await f.write(await r.read())
-
-                    async with aiofiles.open(file_body_path, "w") as f:
-                        await f.write(json.dumps(json.loads(body.decode("utf-8"))))
-
-            return FileResponse(file_path)
-
-        except Exception as e:
-            log.exception(e)
-            error_detail = "Open WebUI: Server Connection Error"
-            try:
-                if r.status != 200:
-                    res = await r.json()
-                    if "error" in res:
-                        error_detail = f"External: {res['error']['message']}"
-            except Exception:
-                error_detail = f"External: {e}"
-
-            raise HTTPException(
-                status_code=getattr(r, "status", 500),
-                detail=error_detail,
-            )
-
-    elif app.state.config.TTS_ENGINE == "azure":
-        try:
-            payload = json.loads(body.decode("utf-8"))
-        except Exception as e:
-            log.exception(e)
-            raise HTTPException(status_code=400, detail="Invalid JSON payload")
-
-        region = app.state.config.TTS_AZURE_SPEECH_REGION
-        language = app.state.config.TTS_VOICE
-        locale = "-".join(app.state.config.TTS_VOICE.split("-")[:1])
-        output_format = app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT
-        url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1"
-
-        headers = {
-            "Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY,
-            "Content-Type": "application/ssml+xml",
-            "X-Microsoft-OutputFormat": output_format,
-        }
-
-        data = f"""<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{locale}">
-                <voice name="{language}">{payload["input"]}</voice>
-            </speak>"""
-
-        try:
-            async with aiohttp.ClientSession() as session:
-                async with session.post(url, headers=headers, data=data) as response:
-                    if response.status == 200:
-                        async with aiofiles.open(file_path, "wb") as f:
-                            await f.write(await response.read())
-                        return FileResponse(file_path)
-                    else:
-                        error_msg = f"Error synthesizing speech - {response.reason}"
-                        log.error(error_msg)
-                        raise HTTPException(status_code=500, detail=error_msg)
-        except Exception as e:
-            log.exception(e)
-            raise HTTPException(status_code=500, detail=str(e))
-    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):
-    print("transcribe", file_path)
-    filename = os.path.basename(file_path)
-    file_dir = os.path.dirname(file_path)
-    id = filename.split(".")[0]
-
-    if app.state.config.STT_ENGINE == "":
-        if app.state.faster_whisper_model is None:
-            set_faster_whisper_model(app.state.config.WHISPER_MODEL)
-
-        model = app.state.faster_whisper_model
-        segments, info = model.transcribe(file_path, beam_size=5)
-        log.info(
-            "Detected language '%s' with probability %f"
-            % (info.language, info.language_probability)
-        )
-
-        transcript = "".join([segment.text for segment in list(segments)])
-        data = {"text": transcript.strip()}
-
-        # save the transcript to a json file
-        transcript_file = f"{file_dir}/{id}.json"
-        with open(transcript_file, "w") as f:
-            json.dump(data, f)
-
-        log.debug(data)
-        return data
-    elif app.state.config.STT_ENGINE == "openai":
-        if is_mp4_audio(file_path):
-            print("is_mp4_audio")
-            os.rename(file_path, file_path.replace(".wav", ".mp4"))
-            # Convert MP4 audio file to WAV format
-            convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path)
-
-        headers = {"Authorization": f"Bearer {app.state.config.STT_OPENAI_API_KEY}"}
-
-        files = {"file": (filename, open(file_path, "rb"))}
-        data = {"model": app.state.config.STT_MODEL}
-
-        log.debug(files, data)
-
-        r = None
-        try:
-            r = requests.post(
-                url=f"{app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions",
-                headers=headers,
-                files=files,
-                data=data,
-            )
-
-            r.raise_for_status()
-
-            data = r.json()
-
-            # save the transcript to a json file
-            transcript_file = f"{file_dir}/{id}.json"
-            with open(transcript_file, "w") as f:
-                json.dump(data, f)
-
-            print(data)
-            return data
-        except Exception as e:
-            log.exception(e)
-            error_detail = "Open WebUI: Server Connection Error"
-            if r is not None:
-                try:
-                    res = r.json()
-                    if "error" in res:
-                        error_detail = f"External: {res['error']['message']}"
-                except Exception:
-                    error_detail = f"External: {e}"
-
-            raise Exception(error_detail)
-
-
-@app.post("/transcriptions")
-def transcription(
-    file: UploadFile = File(...),
-    user=Depends(get_verified_user),
-):
-    log.info(f"file.content_type: {file.content_type}")
-
-    if file.content_type not in ["audio/mpeg", "audio/wav", "audio/ogg", "audio/x-m4a"]:
-        raise HTTPException(
-            status_code=status.HTTP_400_BAD_REQUEST,
-            detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED,
-        )
-
-    try:
-        ext = file.filename.split(".")[-1]
-        id = uuid.uuid4()
-
-        filename = f"{id}.{ext}"
-        contents = file.file.read()
-
-        file_dir = f"{CACHE_DIR}/audio/transcriptions"
-        os.makedirs(file_dir, exist_ok=True)
-        file_path = f"{file_dir}/{filename}"
-
-        with open(file_path, "wb") as f:
-            f.write(contents)
-
-        try:
-            if os.path.getsize(file_path) > MAX_FILE_SIZE:  # file is bigger than 25MB
-                log.debug(f"File size is larger than {MAX_FILE_SIZE_MB}MB")
-                audio = AudioSegment.from_file(file_path)
-                audio = audio.set_frame_rate(16000).set_channels(1)  # Compress audio
-                compressed_path = f"{file_dir}/{id}_compressed.opus"
-                audio.export(compressed_path, format="opus", bitrate="32k")
-                log.debug(f"Compressed audio to {compressed_path}")
-                file_path = compressed_path
-
-                if (
-                    os.path.getsize(file_path) > MAX_FILE_SIZE
-                ):  # Still larger than 25MB after compression
-                    log.debug(
-                        f"Compressed file size is still larger than {MAX_FILE_SIZE_MB}MB: {os.path.getsize(file_path)}"
-                    )
-                    raise HTTPException(
-                        status_code=status.HTTP_400_BAD_REQUEST,
-                        detail=ERROR_MESSAGES.FILE_TOO_LARGE(
-                            size=f"{MAX_FILE_SIZE_MB}MB"
-                        ),
-                    )
-
-                data = transcribe(file_path)
-            else:
-                data = transcribe(file_path)
-
-            file_path = file_path.split("/")[-1]
-            return {**data, "filename": file_path}
-        except Exception as e:
-            log.exception(e)
-            raise HTTPException(
-                status_code=status.HTTP_400_BAD_REQUEST,
-                detail=ERROR_MESSAGES.DEFAULT(e),
-            )
-
-    except Exception as e:
-        log.exception(e)
-
-        raise HTTPException(
-            status_code=status.HTTP_400_BAD_REQUEST,
-            detail=ERROR_MESSAGES.DEFAULT(e),
-        )
-
-
-def get_available_models() -> list[dict]:
-    if app.state.config.TTS_ENGINE == "openai":
-        return [{"id": "tts-1"}, {"id": "tts-1-hd"}]
-    elif app.state.config.TTS_ENGINE == "elevenlabs":
-        headers = {
-            "xi-api-key": app.state.config.TTS_API_KEY,
-            "Content-Type": "application/json",
-        }
-
-        try:
-            response = requests.get(
-                "https://api.elevenlabs.io/v1/models", headers=headers, timeout=5
-            )
-            response.raise_for_status()
-            models = response.json()
-            return [
-                {"name": model["name"], "id": model["model_id"]} for model in models
-            ]
-        except requests.RequestException as e:
-            log.error(f"Error fetching voices: {str(e)}")
-    return []
-
-
-@app.get("/models")
-async def get_models(user=Depends(get_verified_user)):
-    return {"models": get_available_models()}
-
-
-def get_available_voices() -> dict:
-    """Returns {voice_id: voice_name} dict"""
-    ret = {}
-    if app.state.config.TTS_ENGINE == "openai":
-        ret = {
-            "alloy": "alloy",
-            "echo": "echo",
-            "fable": "fable",
-            "onyx": "onyx",
-            "nova": "nova",
-            "shimmer": "shimmer",
-        }
-    elif app.state.config.TTS_ENGINE == "elevenlabs":
-        try:
-            ret = get_elevenlabs_voices()
-        except Exception:
-            # Avoided @lru_cache with exception
-            pass
-    elif app.state.config.TTS_ENGINE == "azure":
-        try:
-            region = app.state.config.TTS_AZURE_SPEECH_REGION
-            url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/voices/list"
-            headers = {"Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY}
-
-            response = requests.get(url, headers=headers)
-            response.raise_for_status()
-            voices = response.json()
-            for voice in voices:
-                ret[voice["ShortName"]] = (
-                    f"{voice['DisplayName']} ({voice['ShortName']})"
-                )
-        except requests.RequestException as e:
-            log.error(f"Error fetching voices: {str(e)}")
-
-    return ret
-
-
-@lru_cache
-def get_elevenlabs_voices() -> dict:
-    """
-    Note, set the following in your .env file to use Elevenlabs:
-    AUDIO_TTS_ENGINE=elevenlabs
-    AUDIO_TTS_API_KEY=sk_...  # Your Elevenlabs API key
-    AUDIO_TTS_VOICE=EXAVITQu4vr4xnSDxMaL  # From https://api.elevenlabs.io/v1/voices
-    AUDIO_TTS_MODEL=eleven_multilingual_v2
-    """
-    headers = {
-        "xi-api-key": app.state.config.TTS_API_KEY,
-        "Content-Type": "application/json",
-    }
-    try:
-        # TODO: Add retries
-        response = requests.get("https://api.elevenlabs.io/v1/voices", headers=headers)
-        response.raise_for_status()
-        voices_data = response.json()
-
-        voices = {}
-        for voice in voices_data.get("voices", []):
-            voices[voice["voice_id"]] = voice["name"]
-    except requests.RequestException as e:
-        # Avoid @lru_cache with exception
-        log.error(f"Error fetching voices: {str(e)}")
-        raise RuntimeError(f"Error fetching voices: {str(e)}")
-
-    return voices
-
-
-@app.get("/voices")
-async def get_voices(user=Depends(get_verified_user)):
-    return {"voices": [{"id": k, "name": v} for k, v in get_available_voices().items()]}

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

@@ -1,22 +0,0 @@
-from open_webui.config import VECTOR_DB
-
-if VECTOR_DB == "milvus":
-    from open_webui.apps.retrieval.vector.dbs.milvus import MilvusClient
-
-    VECTOR_DB_CLIENT = MilvusClient()
-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
-
-    VECTOR_DB_CLIENT = ChromaClient()

backend/open_webui/apps/webui/main.py

@@ -1,506 +0,0 @@
-import inspect
-import json
-import logging
-import time
-from typing import AsyncGenerator, Generator, Iterator
-
-from open_webui.apps.socket.main import get_event_call, get_event_emitter
-from open_webui.apps.webui.models.functions import Functions
-from open_webui.apps.webui.models.models import Models
-from open_webui.apps.webui.routers import (
-    auths,
-    chats,
-    folders,
-    configs,
-    groups,
-    files,
-    functions,
-    memories,
-    models,
-    knowledge,
-    prompts,
-    evaluations,
-    tools,
-    users,
-    utils,
-)
-from open_webui.apps.webui.utils import load_function_module_by_id
-from open_webui.config import (
-    ADMIN_EMAIL,
-    CORS_ALLOW_ORIGIN,
-    DEFAULT_MODELS,
-    DEFAULT_PROMPT_SUGGESTIONS,
-    DEFAULT_USER_ROLE,
-    MODEL_ORDER_LIST,
-    ENABLE_COMMUNITY_SHARING,
-    ENABLE_LOGIN_FORM,
-    ENABLE_MESSAGE_RATING,
-    ENABLE_SIGNUP,
-    ENABLE_API_KEY,
-    ENABLE_EVALUATION_ARENA_MODELS,
-    EVALUATION_ARENA_MODELS,
-    DEFAULT_ARENA_MODEL,
-    JWT_EXPIRES_IN,
-    ENABLE_OAUTH_ROLE_MANAGEMENT,
-    OAUTH_ROLES_CLAIM,
-    OAUTH_EMAIL_CLAIM,
-    OAUTH_PICTURE_CLAIM,
-    OAUTH_USERNAME_CLAIM,
-    OAUTH_ALLOWED_ROLES,
-    OAUTH_ADMIN_ROLES,
-    SHOW_ADMIN_DETAILS,
-    USER_PERMISSIONS,
-    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,
-    SRC_LOG_LEVELS,
-    WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
-    WEBUI_AUTH_TRUSTED_NAME_HEADER,
-)
-from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse
-from pydantic import BaseModel
-from open_webui.utils.misc import (
-    openai_chat_chunk_message_template,
-    openai_chat_completion_message_template,
-)
-from open_webui.utils.payload import (
-    apply_model_params_to_body_openai,
-    apply_model_system_prompt_to_body,
-)
-
-
-from open_webui.utils.tools import get_tools
-
-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__)
-log.setLevel(SRC_LOG_LEVELS["MAIN"])
-
-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
-
-
-app.state.config.SHOW_ADMIN_DETAILS = SHOW_ADMIN_DETAILS
-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
-app.state.config.MODEL_ORDER_LIST = MODEL_ORDER_LIST
-
-app.state.config.ENABLE_COMMUNITY_SHARING = ENABLE_COMMUNITY_SHARING
-app.state.config.ENABLE_MESSAGE_RATING = ENABLE_MESSAGE_RATING
-
-app.state.config.ENABLE_EVALUATION_ARENA_MODELS = ENABLE_EVALUATION_ARENA_MODELS
-app.state.config.EVALUATION_ARENA_MODELS = EVALUATION_ARENA_MODELS
-
-app.state.config.OAUTH_USERNAME_CLAIM = OAUTH_USERNAME_CLAIM
-app.state.config.OAUTH_PICTURE_CLAIM = OAUTH_PICTURE_CLAIM
-app.state.config.OAUTH_EMAIL_CLAIM = OAUTH_EMAIL_CLAIM
-
-app.state.config.ENABLE_OAUTH_ROLE_MANAGEMENT = ENABLE_OAUTH_ROLE_MANAGEMENT
-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.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 = {}
-
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=CORS_ALLOW_ORIGIN,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-
-app.include_router(configs.router, prefix="/configs", tags=["configs"])
-
-app.include_router(auths.router, prefix="/auths", tags=["auths"])
-app.include_router(users.router, prefix="/users", tags=["users"])
-
-app.include_router(chats.router, prefix="/chats", tags=["chats"])
-
-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(memories.router, prefix="/memories", tags=["memories"])
-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"])
-
-
-@app.get("/")
-async def get_status():
-    return {
-        "status": True,
-        "auth": WEBUI_AUTH,
-        "default_models": app.state.config.DEFAULT_MODELS,
-        "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS,
-    }
-
-
-async def get_all_models():
-    models = []
-    pipe_models = await get_pipe_models()
-    models = models + pipe_models
-
-    if app.state.config.ENABLE_EVALUATION_ARENA_MODELS:
-        arena_models = []
-        if len(app.state.config.EVALUATION_ARENA_MODELS) > 0:
-            arena_models = [
-                {
-                    "id": model["id"],
-                    "name": model["name"],
-                    "info": {
-                        "meta": model["meta"],
-                    },
-                    "object": "model",
-                    "created": int(time.time()),
-                    "owned_by": "arena",
-                    "arena": True,
-                }
-                for model in app.state.config.EVALUATION_ARENA_MODELS
-            ]
-        else:
-            # Add default arena model
-            arena_models = [
-                {
-                    "id": DEFAULT_ARENA_MODEL["id"],
-                    "name": DEFAULT_ARENA_MODEL["name"],
-                    "info": {
-                        "meta": DEFAULT_ARENA_MODEL["meta"],
-                    },
-                    "object": "model",
-                    "created": int(time.time()),
-                    "owned_by": "arena",
-                    "arena": True,
-                }
-            ]
-        models = models + arena_models
-    return models
-
-
-def get_function_module(pipe_id: str):
-    # Check if function is already loaded
-    if pipe_id not in app.state.FUNCTIONS:
-        function_module, _, _ = load_function_module_by_id(pipe_id)
-        app.state.FUNCTIONS[pipe_id] = function_module
-    else:
-        function_module = app.state.FUNCTIONS[pipe_id]
-
-    if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
-        valves = Functions.get_function_valves_by_id(pipe_id)
-        function_module.valves = function_module.Valves(**(valves if valves else {}))
-    return function_module
-
-
-async def get_pipe_models():
-    pipes = Functions.get_functions_by_type("pipe", active_only=True)
-    pipe_models = []
-
-    for pipe in pipes:
-        function_module = get_function_module(pipe.id)
-
-        # Check if function is a manifold
-        if hasattr(function_module, "pipes"):
-            sub_pipes = []
-
-            # Check if pipes is a function or a list
-
-            try:
-                if callable(function_module.pipes):
-                    sub_pipes = function_module.pipes()
-                else:
-                    sub_pipes = function_module.pipes
-            except Exception as e:
-                log.exception(e)
-                sub_pipes = []
-
-            log.debug(
-                f"get_pipe_models: function '{pipe.id}' is a manifold of {sub_pipes}"
-            )
-
-            for p in sub_pipes:
-                sub_pipe_id = f'{pipe.id}.{p["id"]}'
-                sub_pipe_name = p["name"]
-
-                if hasattr(function_module, "name"):
-                    sub_pipe_name = f"{function_module.name}{sub_pipe_name}"
-
-                pipe_flag = {"type": pipe.type}
-
-                pipe_models.append(
-                    {
-                        "id": sub_pipe_id,
-                        "name": sub_pipe_name,
-                        "object": "model",
-                        "created": pipe.created_at,
-                        "owned_by": "openai",
-                        "pipe": pipe_flag,
-                    }
-                )
-        else:
-            pipe_flag = {"type": "pipe"}
-
-            log.debug(
-                f"get_pipe_models: function '{pipe.id}' is a single pipe {{ 'id': {pipe.id}, 'name': {pipe.name} }}"
-            )
-
-            pipe_models.append(
-                {
-                    "id": pipe.id,
-                    "name": pipe.name,
-                    "object": "model",
-                    "created": pipe.created_at,
-                    "owned_by": "openai",
-                    "pipe": pipe_flag,
-                }
-            )
-
-    return pipe_models
-
-
-async def execute_pipe(pipe, params):
-    if inspect.iscoroutinefunction(pipe):
-        return await pipe(**params)
-    else:
-        return pipe(**params)
-
-
-async def get_message_content(res: str | Generator | AsyncGenerator) -> str:
-    if isinstance(res, str):
-        return res
-    if isinstance(res, Generator):
-        return "".join(map(str, res))
-    if isinstance(res, AsyncGenerator):
-        return "".join([str(stream) async for stream in res])
-
-
-def process_line(form_data: dict, line):
-    if isinstance(line, BaseModel):
-        line = line.model_dump_json()
-        line = f"data: {line}"
-    if isinstance(line, dict):
-        line = f"data: {json.dumps(line)}"
-
-    try:
-        line = line.decode("utf-8")
-    except Exception:
-        pass
-
-    if line.startswith("data:"):
-        return f"{line}\n\n"
-    else:
-        line = openai_chat_chunk_message_template(form_data["model"], line)
-        return f"data: {json.dumps(line)}\n\n"
-
-
-def get_pipe_id(form_data: dict) -> str:
-    pipe_id = form_data["model"]
-    if "." in pipe_id:
-        pipe_id, _ = pipe_id.split(".", 1)
-
-    return pipe_id
-
-
-def get_function_params(function_module, form_data, user, extra_params=None):
-    if extra_params is None:
-        extra_params = {}
-
-    pipe_id = get_pipe_id(form_data)
-
-    # Get the signature of the function
-    sig = inspect.signature(function_module.pipe)
-    params = {"body": form_data} | {
-        k: v for k, v in extra_params.items() if k in sig.parameters
-    }
-
-    if "__user__" in params and hasattr(function_module, "UserValves"):
-        user_valves = Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id)
-        try:
-            params["__user__"]["valves"] = function_module.UserValves(**user_valves)
-        except Exception as e:
-            log.exception(e)
-            params["__user__"]["valves"] = function_module.UserValves()
-
-    return params
-
-
-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)
-
-    metadata = form_data.pop("metadata", {})
-
-    files = metadata.get("files", [])
-    tool_ids = metadata.get("tool_ids", [])
-    # Check if tool_ids is None
-    if tool_ids is None:
-        tool_ids = []
-
-    __event_emitter__ = None
-    __event_call__ = None
-    __task__ = None
-    __task_body__ = None
-
-    if metadata:
-        if all(k in metadata for k in ("session_id", "chat_id", "message_id")):
-            __event_emitter__ = get_event_emitter(metadata)
-            __event_call__ = get_event_call(metadata)
-        __task__ = metadata.get("task", None)
-        __task_body__ = metadata.get("task_body", None)
-
-    extra_params = {
-        "__event_emitter__": __event_emitter__,
-        "__event_call__": __event_call__,
-        "__task__": __task__,
-        "__task_body__": __task_body__,
-        "__files__": files,
-        "__user__": {
-            "id": user.id,
-            "email": user.email,
-            "name": user.name,
-            "role": user.role,
-        },
-        "__metadata__": metadata,
-    }
-    extra_params["__tools__"] = get_tools(
-        app,
-        tool_ids,
-        user,
-        {
-            **extra_params,
-            "__model__": models.get(form_data["model"], None),
-            "__messages__": form_data["messages"],
-            "__files__": files,
-        },
-    )
-
-    if model_info:
-        if model_info.base_model_id:
-            form_data["model"] = model_info.base_model_id
-
-        params = model_info.params.model_dump()
-        form_data = apply_model_params_to_body_openai(params, form_data)
-        form_data = apply_model_system_prompt_to_body(params, form_data, user)
-
-    pipe_id = get_pipe_id(form_data)
-    function_module = get_function_module(pipe_id)
-
-    pipe = function_module.pipe
-    params = get_function_params(function_module, form_data, user, extra_params)
-
-    if form_data.get("stream", False):
-
-        async def stream_content():
-            try:
-                res = await execute_pipe(pipe, params)
-
-                # Directly return if the response is a StreamingResponse
-                if isinstance(res, StreamingResponse):
-                    async for data in res.body_iterator:
-                        yield data
-                    return
-                if isinstance(res, dict):
-                    yield f"data: {json.dumps(res)}\n\n"
-                    return
-
-            except Exception as e:
-                log.error(f"Error: {e}")
-                yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n"
-                return
-
-            if isinstance(res, str):
-                message = openai_chat_chunk_message_template(form_data["model"], res)
-                yield f"data: {json.dumps(message)}\n\n"
-
-            if isinstance(res, Iterator):
-                for line in res:
-                    yield process_line(form_data, line)
-
-            if isinstance(res, AsyncGenerator):
-                async for line in res:
-                    yield process_line(form_data, line)
-
-            if isinstance(res, str) or isinstance(res, Generator):
-                finish_message = openai_chat_chunk_message_template(
-                    form_data["model"], ""
-                )
-                finish_message["choices"][0]["finish_reason"] = "stop"
-                yield f"data: {json.dumps(finish_message)}\n\n"
-                yield "data: [DONE]"
-
-        return StreamingResponse(stream_content(), media_type="text/event-stream")
-    else:
-        try:
-            res = await execute_pipe(pipe, params)
-
-        except Exception as e:
-            log.error(f"Error: {e}")
-            return {"error": {"detail": str(e)}}
-
-        if isinstance(res, StreamingResponse) or isinstance(res, dict):
-            return res
-        if isinstance(res, BaseModel):
-            return res.model_dump()
-
-        message = await get_message_content(res)
-        return openai_chat_completion_message_template(form_data["model"], message)

backend/open_webui/config.py

@@ -10,7 +10,7 @@ from urllib.parse import urlparse
 import chromadb
 import requests
 import yaml
-from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.internal.db import Base, get_db
 from open_webui.env import (
     OPEN_WEBUI_DIR,
     DATA_DIR,
@@ -21,6 +21,7 @@ from open_webui.env import (
     WEBUI_NAME,
     log,
     DATABASE_URL,
+    OFFLINE_MODE
 )
 from pydantic import BaseModel
 from sqlalchemy import JSON, Column, DateTime, Integer, func
@@ -429,6 +430,15 @@ OAUTH_ADMIN_ROLES = PersistentConfig(
     [role.strip() for role in os.environ.get("OAUTH_ADMIN_ROLES", "admin").split(",")],
 )
 
+OAUTH_ALLOWED_DOMAINS = PersistentConfig(
+    "OAUTH_ALLOWED_DOMAINS",
+    "oauth.allowed_domains",
+    [
+        domain.strip()
+        for domain in os.environ.get("OAUTH_ALLOWED_DOMAINS", "*").split(",")
+    ],
+)
+
 
 def load_oauth_providers():
     OAUTH_PROVIDERS.clear()
@@ -948,12 +958,45 @@ TITLE_GENERATION_PROMPT_TEMPLATE = PersistentConfig(
     os.environ.get("TITLE_GENERATION_PROMPT_TEMPLATE", ""),
 )
 
+DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE = """Create a concise, 3-5 word title with an emoji as a title for the chat history, in the given language. Suitable Emojis for the summary can be used to enhance understanding but avoid quotation marks or special formatting. RESPOND ONLY WITH THE TITLE TEXT.
+
+Examples of titles:
+📉 Stock Market Trends
+🍪 Perfect Chocolate Chip Recipe
+Evolution of Music Streaming
+Remote Work Productivity Tips
+Artificial Intelligence in Healthcare
+🎮 Video Game Development Insights
+
+<chat_history>
+{{MESSAGES:END:2}}
+</chat_history>"""
+
+
 TAGS_GENERATION_PROMPT_TEMPLATE = PersistentConfig(
     "TAGS_GENERATION_PROMPT_TEMPLATE",
     "task.tags.prompt_template",
     os.environ.get("TAGS_GENERATION_PROMPT_TEMPLATE", ""),
 )
 
+DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE = """### Task:
+Generate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags.
+
+### Guidelines:
+- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education)
+- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation
+- If content is too short (less than 3 messages) or too diverse, use only ["General"]
+- Use the chat's primary language; default to English if multilingual
+- Prioritize accuracy over specificity
+
+### Output:
+JSON format: { "tags": ["tag1", "tag2", "tag3"] }
+
+### Chat History:
+<chat_history>
+{{MESSAGES:END:6}}
+</chat_history>"""
+
 ENABLE_TAGS_GENERATION = PersistentConfig(
     "ENABLE_TAGS_GENERATION",
     "task.tags.enable",
@@ -1072,6 +1115,19 @@ TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = PersistentConfig(
 )
 
 
+DEFAULT_TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = """Available Tools: {{TOOLS}}\nReturn an empty string if no tools match the query. If a function tool matches, construct and return a JSON object in the format {\"name\": \"functionName\", \"parameters\": {\"requiredFunctionParamKey\": \"requiredFunctionParamValue\"}} using the appropriate tool and its parameters. Only return the object and limit the response to the JSON object without additional text."""
+
+
+DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE = """Your task is to reflect the speaker's likely facial expression through a fitting emoji. Interpret emotions from the message and reflect their facial expression using fitting, diverse emojis (e.g., 😊, 😢, 😡, 😱).
+
+Message: ```{{prompt}}```"""
+
+DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE = """You have been provided with a set of responses from various models to the latest user query: "{{prompt}}"
+
+Your task is to synthesize these responses into a single, high-quality response. It is crucial to critically evaluate the information provided in these responses, recognizing that some of it may be biased or incorrect. Your response should not simply replicate the given answers but should offer a refined, accurate, and comprehensive reply to the instruction. Ensure your response is well-structured, coherent, and adheres to the highest standards of accuracy and reliability.
+
+Responses from models: {{responses}}"""
+
 ####################################
 # Vector Database
 ####################################
@@ -1197,7 +1253,7 @@ 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", "True").lower() == "true"
+    not OFFLINE_MODE and os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "True").lower() == "true"
 )
 
 RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE = (
@@ -1222,7 +1278,7 @@ 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", "True").lower() == "true"
+    not OFFLINE_MODE and os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "True").lower() == "true"
 )
 
 RAG_RERANKING_MODEL_TRUST_REMOTE_CODE = (
@@ -1380,6 +1436,12 @@ BRAVE_SEARCH_API_KEY = PersistentConfig(
     os.getenv("BRAVE_SEARCH_API_KEY", ""),
 )
 
+KAGI_SEARCH_API_KEY = PersistentConfig(
+    "KAGI_SEARCH_API_KEY",
+    "rag.web.search.kagi_search_api_key",
+    os.getenv("KAGI_SEARCH_API_KEY", ""),
+)
+
 MOJEEK_SEARCH_API_KEY = PersistentConfig(
     "MOJEEK_SEARCH_API_KEY",
     "rag.web.search.mojeek_search_api_key",
@@ -1686,7 +1748,7 @@ WHISPER_MODEL = PersistentConfig(
 
 WHISPER_MODEL_DIR = os.getenv("WHISPER_MODEL_DIR", f"{CACHE_DIR}/whisper/models")
 WHISPER_MODEL_AUTO_UPDATE = (
-    os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true"
+    not OFFLINE_MODE and os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true"
 )
 
 

backend/open_webui/env.py

@@ -378,7 +378,7 @@ else:
         AIOHTTP_CLIENT_TIMEOUT = 300
 
 AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST = os.environ.get(
-    "AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST", "5"
+    "AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST", ""
 )
 
 if AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST == "":

backend/open_webui/functions.py

@@ -0,0 +1,316 @@
+import logging
+import sys
+import inspect
+import json
+
+from pydantic import BaseModel
+from typing import AsyncGenerator, Generator, Iterator
+from fastapi import (
+    Depends,
+    FastAPI,
+    File,
+    Form,
+    HTTPException,
+    Request,
+    UploadFile,
+    status,
+)
+from starlette.responses import Response, StreamingResponse
+
+
+from open_webui.socket.main import (
+    get_event_call,
+    get_event_emitter,
+)
+
+
+from open_webui.models.functions import Functions
+from open_webui.models.models import Models
+
+from open_webui.utils.plugin import load_function_module_by_id
+from open_webui.utils.tools import get_tools
+from open_webui.utils.access_control import has_access
+
+from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL
+
+from open_webui.utils.misc import (
+    add_or_update_system_message,
+    get_last_user_message,
+    prepend_to_first_user_message_content,
+    openai_chat_chunk_message_template,
+    openai_chat_completion_message_template,
+)
+from open_webui.utils.payload import (
+    apply_model_params_to_body_openai,
+    apply_model_system_prompt_to_body,
+)
+
+
+logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL)
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MAIN"])
+
+
+def get_function_module_by_id(request: Request, pipe_id: str):
+    # Check if function is already loaded
+    if pipe_id not in request.app.state.FUNCTIONS:
+        function_module, _, _ = load_function_module_by_id(pipe_id)
+        request.app.state.FUNCTIONS[pipe_id] = function_module
+    else:
+        function_module = request.app.state.FUNCTIONS[pipe_id]
+
+    if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
+        valves = Functions.get_function_valves_by_id(pipe_id)
+        function_module.valves = function_module.Valves(**(valves if valves else {}))
+    return function_module
+
+
+async def get_function_models(request):
+    pipes = Functions.get_functions_by_type("pipe", active_only=True)
+    pipe_models = []
+
+    for pipe in pipes:
+        function_module = get_function_module_by_id(request, pipe.id)
+
+        # Check if function is a manifold
+        if hasattr(function_module, "pipes"):
+            sub_pipes = []
+
+            # Check if pipes is a function or a list
+
+            try:
+                if callable(function_module.pipes):
+                    sub_pipes = function_module.pipes()
+                else:
+                    sub_pipes = function_module.pipes
+            except Exception as e:
+                log.exception(e)
+                sub_pipes = []
+
+            log.debug(
+                f"get_function_models: function '{pipe.id}' is a manifold of {sub_pipes}"
+            )
+
+            for p in sub_pipes:
+                sub_pipe_id = f'{pipe.id}.{p["id"]}'
+                sub_pipe_name = p["name"]
+
+                if hasattr(function_module, "name"):
+                    sub_pipe_name = f"{function_module.name}{sub_pipe_name}"
+
+                pipe_flag = {"type": pipe.type}
+
+                pipe_models.append(
+                    {
+                        "id": sub_pipe_id,
+                        "name": sub_pipe_name,
+                        "object": "model",
+                        "created": pipe.created_at,
+                        "owned_by": "openai",
+                        "pipe": pipe_flag,
+                    }
+                )
+        else:
+            pipe_flag = {"type": "pipe"}
+
+            log.debug(
+                f"get_function_models: function '{pipe.id}' is a single pipe {{ 'id': {pipe.id}, 'name': {pipe.name} }}"
+            )
+
+            pipe_models.append(
+                {
+                    "id": pipe.id,
+                    "name": pipe.name,
+                    "object": "model",
+                    "created": pipe.created_at,
+                    "owned_by": "openai",
+                    "pipe": pipe_flag,
+                }
+            )
+
+    return pipe_models
+
+
+async def generate_function_chat_completion(
+    request, form_data, user, models: dict = {}
+):
+    async def execute_pipe(pipe, params):
+        if inspect.iscoroutinefunction(pipe):
+            return await pipe(**params)
+        else:
+            return pipe(**params)
+
+    async def get_message_content(res: str | Generator | AsyncGenerator) -> str:
+        if isinstance(res, str):
+            return res
+        if isinstance(res, Generator):
+            return "".join(map(str, res))
+        if isinstance(res, AsyncGenerator):
+            return "".join([str(stream) async for stream in res])
+
+    def process_line(form_data: dict, line):
+        if isinstance(line, BaseModel):
+            line = line.model_dump_json()
+            line = f"data: {line}"
+        if isinstance(line, dict):
+            line = f"data: {json.dumps(line)}"
+
+        try:
+            line = line.decode("utf-8")
+        except Exception:
+            pass
+
+        if line.startswith("data:"):
+            return f"{line}\n\n"
+        else:
+            line = openai_chat_chunk_message_template(form_data["model"], line)
+            return f"data: {json.dumps(line)}\n\n"
+
+    def get_pipe_id(form_data: dict) -> str:
+        pipe_id = form_data["model"]
+        if "." in pipe_id:
+            pipe_id, _ = pipe_id.split(".", 1)
+        return pipe_id
+
+    def get_function_params(function_module, form_data, user, extra_params=None):
+        if extra_params is None:
+            extra_params = {}
+
+        pipe_id = get_pipe_id(form_data)
+
+        # Get the signature of the function
+        sig = inspect.signature(function_module.pipe)
+        params = {"body": form_data} | {
+            k: v for k, v in extra_params.items() if k in sig.parameters
+        }
+
+        if "__user__" in params and hasattr(function_module, "UserValves"):
+            user_valves = Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id)
+            try:
+                params["__user__"]["valves"] = function_module.UserValves(**user_valves)
+            except Exception as e:
+                log.exception(e)
+                params["__user__"]["valves"] = function_module.UserValves()
+
+        return params
+
+    model_id = form_data.get("model")
+    model_info = Models.get_model_by_id(model_id)
+
+    metadata = form_data.pop("metadata", {})
+
+    files = metadata.get("files", [])
+    tool_ids = metadata.get("tool_ids", [])
+    # Check if tool_ids is None
+    if tool_ids is None:
+        tool_ids = []
+
+    __event_emitter__ = None
+    __event_call__ = None
+    __task__ = None
+    __task_body__ = None
+
+    if metadata:
+        if all(k in metadata for k in ("session_id", "chat_id", "message_id")):
+            __event_emitter__ = get_event_emitter(metadata)
+            __event_call__ = get_event_call(metadata)
+        __task__ = metadata.get("task", None)
+        __task_body__ = metadata.get("task_body", None)
+
+    extra_params = {
+        "__event_emitter__": __event_emitter__,
+        "__event_call__": __event_call__,
+        "__task__": __task__,
+        "__task_body__": __task_body__,
+        "__files__": files,
+        "__user__": {
+            "id": user.id,
+            "email": user.email,
+            "name": user.name,
+            "role": user.role,
+        },
+        "__metadata__": metadata,
+        "__request__": request,
+    }
+    extra_params["__tools__"] = get_tools(
+        request,
+        tool_ids,
+        user,
+        {
+            **extra_params,
+            "__model__": models.get(form_data["model"], None),
+            "__messages__": form_data["messages"],
+            "__files__": files,
+        },
+    )
+
+    if model_info:
+        if model_info.base_model_id:
+            form_data["model"] = model_info.base_model_id
+
+        params = model_info.params.model_dump()
+        form_data = apply_model_params_to_body_openai(params, form_data)
+        form_data = apply_model_system_prompt_to_body(params, form_data, user)
+
+    pipe_id = get_pipe_id(form_data)
+    function_module = get_function_module_by_id(request, pipe_id)
+
+    pipe = function_module.pipe
+    params = get_function_params(function_module, form_data, user, extra_params)
+
+    if form_data.get("stream", False):
+
+        async def stream_content():
+            try:
+                res = await execute_pipe(pipe, params)
+
+                # Directly return if the response is a StreamingResponse
+                if isinstance(res, StreamingResponse):
+                    async for data in res.body_iterator:
+                        yield data
+                    return
+                if isinstance(res, dict):
+                    yield f"data: {json.dumps(res)}\n\n"
+                    return
+
+            except Exception as e:
+                log.error(f"Error: {e}")
+                yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n"
+                return
+
+            if isinstance(res, str):
+                message = openai_chat_chunk_message_template(form_data["model"], res)
+                yield f"data: {json.dumps(message)}\n\n"
+
+            if isinstance(res, Iterator):
+                for line in res:
+                    yield process_line(form_data, line)
+
+            if isinstance(res, AsyncGenerator):
+                async for line in res:
+                    yield process_line(form_data, line)
+
+            if isinstance(res, str) or isinstance(res, Generator):
+                finish_message = openai_chat_chunk_message_template(
+                    form_data["model"], ""
+                )
+                finish_message["choices"][0]["finish_reason"] = "stop"
+                yield f"data: {json.dumps(finish_message)}\n\n"
+                yield "data: [DONE]"
+
+        return StreamingResponse(stream_content(), media_type="text/event-stream")
+    else:
+        try:
+            res = await execute_pipe(pipe, params)
+
+        except Exception as e:
+            log.error(f"Error: {e}")
+            return {"error": {"detail": str(e)}}
+
+        if isinstance(res, StreamingResponse) or isinstance(res, dict):
+            return res
+        if isinstance(res, BaseModel):
+            return res.model_dump()
+
+        message = await get_message_content(res)
+        return openai_chat_completion_message_template(form_data["model"], message)

backend/open_webui/apps/webui/internal/db.py → backend/open_webui/internal/db.py

@@ -3,7 +3,7 @@ import logging
 from contextlib import contextmanager
 from typing import Any, Optional
 
-from open_webui.apps.webui.internal.wrappers import register_connection
+from open_webui.internal.wrappers import register_connection
 from open_webui.env import (
     OPEN_WEBUI_DIR,
     DATABASE_URL,

backend/open_webui/migrations/env.py

@@ -1,7 +1,7 @@
 from logging.config import fileConfig
 
 from alembic import context
-from open_webui.apps.webui.models.auths import Auth
+from open_webui.models.auths import Auth
 from open_webui.env import DATABASE_URL
 from sqlalchemy import engine_from_config, pool
 

backend/open_webui/migrations/script.py.mako

@@ -9,7 +9,7 @@ from typing import Sequence, Union
 
 from alembic import op
 import sqlalchemy as sa
-import open_webui.apps.webui.internal.db
+import open_webui.internal.db
 ${imports if imports else ""}
 
 # revision identifiers, used by Alembic.

backend/open_webui/migrations/versions/7e5b5dc7342b_init.py

@@ -11,8 +11,8 @@ from typing import Sequence, Union
 import sqlalchemy as sa
 from alembic import op
 
-import open_webui.apps.webui.internal.db
-from open_webui.apps.webui.internal.db import JSONField
+import open_webui.internal.db
+from open_webui.internal.db import JSONField
 from open_webui.migrations.util import get_existing_tables
 
 # revision identifiers, used by Alembic.

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

@@ -2,12 +2,12 @@ import logging
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
-from open_webui.apps.webui.models.users import UserModel, Users
+from open_webui.internal.db import Base, get_db
+from open_webui.models.users import UserModel, Users
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel
 from sqlalchemy import Boolean, Column, String, Text
-from open_webui.utils.utils import verify_password
+from open_webui.utils.auth import verify_password
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])

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

@@ -3,8 +3,8 @@ import time
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
-from open_webui.apps.webui.models.tags import TagModel, Tag, Tags
+from open_webui.internal.db import Base, get_db
+from open_webui.models.tags import TagModel, Tag, Tags
 
 
 from pydantic import BaseModel, ConfigDict

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

@@ -3,8 +3,8 @@ import time
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
-from open_webui.apps.webui.models.chats import Chats
+from open_webui.internal.db import Base, get_db
+from open_webui.models.chats import Chats
 
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict

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

@@ -2,7 +2,7 @@ import logging
 import time
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.internal.db import Base, JSONField, get_db
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Column, String, Text, JSON

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

@@ -3,8 +3,8 @@ import time
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
-from open_webui.apps.webui.models.chats import Chats
+from open_webui.internal.db import Base, get_db
+from open_webui.models.chats import Chats
 
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict

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

@@ -2,8 +2,8 @@ import logging
 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.internal.db import Base, JSONField, get_db
+from open_webui.models.users import Users
 from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Boolean, Column, String, Text

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

@@ -4,10 +4,10 @@ import time
 from typing import Optional
 import uuid
 
-from open_webui.apps.webui.internal.db import Base, get_db
+from open_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.models.files import FileMetadataResponse
 
 
 from pydantic import BaseModel, ConfigDict

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

@@ -4,11 +4,11 @@ import time
 from typing import Optional
 import uuid
 
-from open_webui.apps.webui.internal.db import Base, get_db
+from open_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 open_webui.models.files import FileMetadataResponse
+from open_webui.models.users import Users, UserResponse
 
 
 from pydantic import BaseModel, ConfigDict

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

@@ -2,7 +2,7 @@ import time
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.internal.db import Base, get_db
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Column, String, Text
 

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

@@ -2,10 +2,10 @@ import logging
 import time
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_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 open_webui.models.users import Users, UserResponse
 
 
 from pydantic import BaseModel, ConfigDict

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

@@ -1,8 +1,8 @@
 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 open_webui.internal.db import Base, get_db
+from open_webui.models.users import Users, UserResponse
 
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Column, String, Text, JSON

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

@@ -3,7 +3,7 @@ import time
 import uuid
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.internal.db import Base, get_db
 
 
 from open_webui.env import SRC_LOG_LEVELS

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

@@ -2,8 +2,8 @@ import logging
 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, UserResponse
+from open_webui.internal.db import Base, JSONField, get_db
+from open_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, JSON

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

@@ -1,8 +1,8 @@
 import time
 from typing import Optional
 
-from open_webui.apps.webui.internal.db import Base, JSONField, get_db
-from open_webui.apps.webui.models.chats import Chats
+from open_webui.internal.db import Base, JSONField, get_db
+from open_webui.models.chats import Chats
 from pydantic import BaseModel, ConfigDict
 from sqlalchemy import BigInteger, Column, String, Text
 

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

@@ -1,6 +1,7 @@
 import requests
 import logging
 import ftfy
+import sys
 
 from langchain_community.document_loaders import (
     BSHTMLLoader,
@@ -18,8 +19,9 @@ from langchain_community.document_loaders import (
     YoutubeLoader,
 )
 from langchain_core.documents import Document
-from open_webui.env import SRC_LOG_LEVELS
+from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL
 
+logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL)
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
@@ -106,7 +108,7 @@ class TikaLoader:
             if "Content-Type" in raw_metadata:
                 headers["Content-Type"] = raw_metadata["Content-Type"]
 
-            log.info("Tika extracted text: %s", text)
+            log.debug("Tika extracted text: %s", text)
 
             return [Document(page_content=text, metadata=headers)]
         else:

backend/open_webui/apps/retrieval/utils.py → backend/open_webui/retrieval/utils.py

@@ -11,7 +11,7 @@ from langchain.retrievers import ContextualCompressionRetriever, EnsembleRetriev
 from langchain_community.retrievers import BM25Retriever
 from langchain_core.documents import Document
 
-from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
+from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT
 from open_webui.utils.misc import get_last_user_message
 
 from open_webui.env import SRC_LOG_LEVELS

backend/open_webui/retrieval/vector/connector.py

@@ -0,0 +1,22 @@
+from open_webui.config import VECTOR_DB
+
+if VECTOR_DB == "milvus":
+    from open_webui.retrieval.vector.dbs.milvus import MilvusClient
+
+    VECTOR_DB_CLIENT = MilvusClient()
+elif VECTOR_DB == "qdrant":
+    from open_webui.retrieval.vector.dbs.qdrant import QdrantClient
+
+    VECTOR_DB_CLIENT = QdrantClient()
+elif VECTOR_DB == "opensearch":
+    from open_webui.retrieval.vector.dbs.opensearch import OpenSearchClient
+
+    VECTOR_DB_CLIENT = OpenSearchClient()
+elif VECTOR_DB == "pgvector":
+    from open_webui.retrieval.vector.dbs.pgvector import PgvectorClient
+
+    VECTOR_DB_CLIENT = PgvectorClient()
+else:
+    from open_webui.retrieval.vector.dbs.chroma import ChromaClient
+
+    VECTOR_DB_CLIENT = ChromaClient()

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

@@ -4,7 +4,7 @@ from chromadb.utils.batch_utils import create_batches
 
 from typing import Optional
 
-from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult
 from open_webui.config import (
     CHROMA_DATA_PATH,
     CHROMA_HTTP_HOST,

backend/open_webui/apps/retrieval/vector/dbs/milvus.py → backend/open_webui/retrieval/vector/dbs/milvus.py

@@ -4,7 +4,7 @@ import json
 
 from typing import Optional
 
-from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult
 from open_webui.config import (
     MILVUS_URI,
 )

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

@@ -1,7 +1,7 @@
 from opensearchpy import OpenSearch
 from typing import Optional
 
-from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult
 from open_webui.config import (
     OPENSEARCH_URI,
     OPENSEARCH_SSL,

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

@@ -18,7 +18,7 @@ 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.retrieval.vector.main import VectorItem, SearchResult, GetResult
 from open_webui.config import PGVECTOR_DB_URL
 
 VECTOR_LENGTH = 1536
@@ -40,7 +40,7 @@ class PgvectorClient:
 
         # if no pgvector uri, use the existing database connection
         if not PGVECTOR_DB_URL:
-            from open_webui.apps.webui.internal.db import Session
+            from open_webui.internal.db import Session
 
             self.session = Session
         else:

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

@@ -4,7 +4,7 @@ from qdrant_client import QdrantClient as Qclient
 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.retrieval.vector.main import VectorItem, SearchResult, GetResult
 from open_webui.config import QDRANT_URI, QDRANT_API_KEY
 
 NO_LIMIT = 999999999

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

@@ -3,7 +3,7 @@ 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.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 import argparse
 

backend/open_webui/apps/retrieval/web/brave.py → backend/open_webui/retrieval/web/brave.py

@@ -2,7 +2,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/duckduckgo.py → backend/open_webui/retrieval/web/duckduckgo.py

@@ -1,7 +1,7 @@
 import logging
 from typing import Optional
 
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from duckduckgo_search import DDGS
 from open_webui.env import SRC_LOG_LEVELS
 

backend/open_webui/apps/retrieval/web/google_pse.py → backend/open_webui/retrieval/web/google_pse.py

@@ -2,7 +2,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

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

@@ -1,7 +1,7 @@
 import logging
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult
+from open_webui.retrieval.web.main import SearchResult
 from open_webui.env import SRC_LOG_LEVELS
 from yarl import URL
 

backend/open_webui/retrieval/web/kagi.py

@@ -0,0 +1,48 @@
+import logging
+from typing import Optional
+
+import requests
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["RAG"])
+
+
+def search_kagi(
+    api_key: str, query: str, count: int, filter_list: Optional[list[str]] = None
+) -> list[SearchResult]:
+    """Search using Kagi's Search API and return the results as a list of SearchResult objects.
+
+    The Search API will inherit the settings in your account, including results personalization and snippet length.
+
+    Args:
+        api_key (str): A Kagi Search API key
+        query (str): The query to search for
+        count (int): The number of results to return
+    """
+    url = "https://kagi.com/api/v0/search"
+    headers = {
+        "Authorization": f"Bot {api_key}",
+    }
+    params = {"q": query, "limit": count}
+
+    response = requests.get(url, headers=headers, params=params)
+    response.raise_for_status()
+    json_response = response.json()
+    search_results = json_response.get("data", [])
+
+    results = [
+        SearchResult(
+            link=result["url"], title=result["title"], snippet=result.get("snippet")
+        )
+        for result in search_results
+        if result["t"] == 0
+    ]
+
+    print(results)
+
+    if filter_list:
+        results = get_filtered_results(results, filter_list)
+
+    return results

backend/open_webui/apps/retrieval/web/mojeek.py → backend/open_webui/retrieval/web/mojeek.py

@@ -2,7 +2,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/searchapi.py → backend/open_webui/retrieval/web/searchapi.py

@@ -3,7 +3,7 @@ from typing import Optional
 from urllib.parse import urlencode
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/searxng.py → backend/open_webui/retrieval/web/searxng.py

@@ -2,7 +2,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/serper.py → backend/open_webui/retrieval/web/serper.py

@@ -3,7 +3,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/serply.py → backend/open_webui/retrieval/web/serply.py

@@ -3,7 +3,7 @@ from typing import Optional
 from urllib.parse import urlencode
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/serpstack.py → backend/open_webui/retrieval/web/serpstack.py

@@ -2,7 +2,7 @@ import logging
 from typing import Optional
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results
+from open_webui.retrieval.web.main import SearchResult, get_filtered_results
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/apps/retrieval/web/tavily.py → backend/open_webui/retrieval/web/tavily.py

@@ -1,7 +1,7 @@
 import logging
 
 import requests
-from open_webui.apps.retrieval.web.main import SearchResult
+from open_webui.retrieval.web.main import SearchResult
 from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)

backend/open_webui/routers/audio.py

@@ -0,0 +1,703 @@
+import hashlib
+import json
+import logging
+import os
+import uuid
+from functools import lru_cache
+from pathlib import Path
+from pydub import AudioSegment
+from pydub.silence import split_on_silence
+
+import aiohttp
+import aiofiles
+import requests
+
+from fastapi import (
+    Depends,
+    FastAPI,
+    File,
+    HTTPException,
+    Request,
+    UploadFile,
+    status,
+    APIRouter,
+)
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import FileResponse
+from pydantic import BaseModel
+
+
+from open_webui.utils.auth import get_admin_user, get_verified_user
+from open_webui.config import (
+    WHISPER_MODEL_AUTO_UPDATE,
+    WHISPER_MODEL_DIR,
+    CACHE_DIR,
+)
+
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import (
+    ENV,
+    SRC_LOG_LEVELS,
+    DEVICE_TYPE,
+    ENABLE_FORWARD_USER_INFO_HEADERS,
+)
+
+
+router = APIRouter()
+
+# Constants
+MAX_FILE_SIZE_MB = 25
+MAX_FILE_SIZE = MAX_FILE_SIZE_MB * 1024 * 1024  # Convert MB to bytes
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["AUDIO"])
+
+SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
+SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
+
+
+##########################################
+#
+# Utility functions
+#
+##########################################
+
+from pydub import AudioSegment
+from pydub.utils import mediainfo
+
+
+def is_mp4_audio(file_path):
+    """Check if the given file is an MP4 audio file."""
+    if not os.path.isfile(file_path):
+        print(f"File not found: {file_path}")
+        return False
+
+    info = mediainfo(file_path)
+    if (
+        info.get("codec_name") == "aac"
+        and info.get("codec_type") == "audio"
+        and info.get("codec_tag_string") == "mp4a"
+    ):
+        return True
+    return False
+
+
+def convert_mp4_to_wav(file_path, output_path):
+    """Convert MP4 audio file to WAV format."""
+    audio = AudioSegment.from_file(file_path, format="mp4")
+    audio.export(output_path, format="wav")
+    print(f"Converted {file_path} to {output_path}")
+
+
+def set_faster_whisper_model(model: str, auto_update: bool = False):
+    whisper_model = None
+    if model:
+        from faster_whisper import WhisperModel
+
+        faster_whisper_kwargs = {
+            "model_size_or_path": model,
+            "device": DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu",
+            "compute_type": "int8",
+            "download_root": WHISPER_MODEL_DIR,
+            "local_files_only": not auto_update,
+        }
+
+        try:
+            whisper_model = WhisperModel(**faster_whisper_kwargs)
+        except Exception:
+            log.warning(
+                "WhisperModel initialization failed, attempting download with local_files_only=False"
+            )
+            faster_whisper_kwargs["local_files_only"] = False
+            whisper_model = WhisperModel(**faster_whisper_kwargs)
+    return whisper_model
+
+
+##########################################
+#
+# Audio API
+#
+##########################################
+
+
+class TTSConfigForm(BaseModel):
+    OPENAI_API_BASE_URL: str
+    OPENAI_API_KEY: str
+    API_KEY: str
+    ENGINE: str
+    MODEL: str
+    VOICE: str
+    SPLIT_ON: str
+    AZURE_SPEECH_REGION: str
+    AZURE_SPEECH_OUTPUT_FORMAT: str
+
+
+class STTConfigForm(BaseModel):
+    OPENAI_API_BASE_URL: str
+    OPENAI_API_KEY: str
+    ENGINE: str
+    MODEL: str
+    WHISPER_MODEL: str
+
+
+class AudioConfigUpdateForm(BaseModel):
+    tts: TTSConfigForm
+    stt: STTConfigForm
+
+
+@router.get("/config")
+async def get_audio_config(request: Request, user=Depends(get_admin_user)):
+    return {
+        "tts": {
+            "OPENAI_API_BASE_URL": request.app.state.config.TTS_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.TTS_OPENAI_API_KEY,
+            "API_KEY": request.app.state.config.TTS_API_KEY,
+            "ENGINE": request.app.state.config.TTS_ENGINE,
+            "MODEL": request.app.state.config.TTS_MODEL,
+            "VOICE": request.app.state.config.TTS_VOICE,
+            "SPLIT_ON": request.app.state.config.TTS_SPLIT_ON,
+            "AZURE_SPEECH_REGION": request.app.state.config.TTS_AZURE_SPEECH_REGION,
+            "AZURE_SPEECH_OUTPUT_FORMAT": request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
+        },
+        "stt": {
+            "OPENAI_API_BASE_URL": request.app.state.config.STT_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.STT_OPENAI_API_KEY,
+            "ENGINE": request.app.state.config.STT_ENGINE,
+            "MODEL": request.app.state.config.STT_MODEL,
+            "WHISPER_MODEL": request.app.state.config.WHISPER_MODEL,
+        },
+    }
+
+
+@router.post("/config/update")
+async def update_audio_config(
+    request: Request, form_data: AudioConfigUpdateForm, user=Depends(get_admin_user)
+):
+    request.app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL
+    request.app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY
+    request.app.state.config.TTS_API_KEY = form_data.tts.API_KEY
+    request.app.state.config.TTS_ENGINE = form_data.tts.ENGINE
+    request.app.state.config.TTS_MODEL = form_data.tts.MODEL
+    request.app.state.config.TTS_VOICE = form_data.tts.VOICE
+    request.app.state.config.TTS_SPLIT_ON = form_data.tts.SPLIT_ON
+    request.app.state.config.TTS_AZURE_SPEECH_REGION = form_data.tts.AZURE_SPEECH_REGION
+    request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = (
+        form_data.tts.AZURE_SPEECH_OUTPUT_FORMAT
+    )
+
+    request.app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL
+    request.app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY
+    request.app.state.config.STT_ENGINE = form_data.stt.ENGINE
+    request.app.state.config.STT_MODEL = form_data.stt.MODEL
+    request.app.state.config.WHISPER_MODEL = form_data.stt.WHISPER_MODEL
+
+    if request.app.state.config.STT_ENGINE == "":
+        request.app.state.faster_whisper_model = set_faster_whisper_model(
+            form_data.stt.WHISPER_MODEL, WHISPER_MODEL_AUTO_UPDATE
+        )
+
+    return {
+        "tts": {
+            "OPENAI_API_BASE_URL": request.app.state.config.TTS_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.TTS_OPENAI_API_KEY,
+            "API_KEY": request.app.state.config.TTS_API_KEY,
+            "ENGINE": request.app.state.config.TTS_ENGINE,
+            "MODEL": request.app.state.config.TTS_MODEL,
+            "VOICE": request.app.state.config.TTS_VOICE,
+            "SPLIT_ON": request.app.state.config.TTS_SPLIT_ON,
+            "AZURE_SPEECH_REGION": request.app.state.config.TTS_AZURE_SPEECH_REGION,
+            "AZURE_SPEECH_OUTPUT_FORMAT": request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
+        },
+        "stt": {
+            "OPENAI_API_BASE_URL": request.app.state.config.STT_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.STT_OPENAI_API_KEY,
+            "ENGINE": request.app.state.config.STT_ENGINE,
+            "MODEL": request.app.state.config.STT_MODEL,
+            "WHISPER_MODEL": request.app.state.config.WHISPER_MODEL,
+        },
+    }
+
+
+def load_speech_pipeline():
+    from transformers import pipeline
+    from datasets import load_dataset
+
+    if request.app.state.speech_synthesiser is None:
+        request.app.state.speech_synthesiser = pipeline(
+            "text-to-speech", "microsoft/speecht5_tts"
+        )
+
+    if request.app.state.speech_speaker_embeddings_dataset is None:
+        request.app.state.speech_speaker_embeddings_dataset = load_dataset(
+            "Matthijs/cmu-arctic-xvectors", split="validation"
+        )
+
+
+@router.post("/speech")
+async def speech(request: Request, user=Depends(get_verified_user)):
+    body = await request.body()
+    name = hashlib.sha256(body).hexdigest()
+
+    file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
+    file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
+
+    # Check if the file already exists in the cache
+    if file_path.is_file():
+        return FileResponse(file_path)
+
+    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")
+
+    if request.app.state.config.TTS_ENGINE == "openai":
+        payload["model"] = request.app.state.config.TTS_MODEL
+
+        try:
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    url=f"{request.app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech",
+                    data=payload,
+                    headers={
+                        "Content-Type": "application/json",
+                        "Authorization": f"Bearer {request.app.state.config.TTS_OPENAI_API_KEY}",
+                        **(
+                            {
+                                "X-OpenWebUI-User-Name": user.name,
+                                "X-OpenWebUI-User-Id": user.id,
+                                "X-OpenWebUI-User-Email": user.email,
+                                "X-OpenWebUI-User-Role": user.role,
+                            }
+                            if ENABLE_FORWARD_USER_INFO_HEADERS
+                            else {}
+                        ),
+                    },
+                ) as r:
+                    r.raise_for_status()
+
+                    async with aiofiles.open(file_path, "wb") as f:
+                        await f.write(await r.read())
+
+                    async with aiofiles.open(file_body_path, "w") as f:
+                        await f.write(json.dumps(json.loads(body.decode("utf-8"))))
+
+            return FileResponse(file_path)
+
+        except Exception as e:
+            log.exception(e)
+            detail = None
+
+            try:
+                if r.status != 200:
+                    res = await r.json()
+                    if "error" in res:
+                        detail = f"External: {res['error'].get('message', '')}"
+            except Exception:
+                detail = f"External: {e}"
+
+            raise HTTPException(
+                status_code=getattr(r, "status", 500),
+                detail=detail if detail else "Open WebUI: Server Connection Error",
+            )
+
+    elif request.app.state.config.TTS_ENGINE == "elevenlabs":
+        voice_id = payload.get("voice", "")
+
+        if voice_id not in get_available_voices():
+            raise HTTPException(
+                status_code=400,
+                detail="Invalid voice id",
+            )
+
+        try:
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}",
+                    json={
+                        "text": payload["input"],
+                        "model_id": request.app.state.config.TTS_MODEL,
+                        "voice_settings": {"stability": 0.5, "similarity_boost": 0.5},
+                    },
+                    headers={
+                        "Accept": "audio/mpeg",
+                        "Content-Type": "application/json",
+                        "xi-api-key": request.app.state.config.TTS_API_KEY,
+                    },
+                ) as r:
+                    r.raise_for_status()
+
+                    async with aiofiles.open(file_path, "wb") as f:
+                        await f.write(await r.read())
+
+                    async with aiofiles.open(file_body_path, "w") as f:
+                        await f.write(json.dumps(json.loads(body.decode("utf-8"))))
+
+            return FileResponse(file_path)
+
+        except Exception as e:
+            log.exception(e)
+            detail = None
+
+            try:
+                if r.status != 200:
+                    res = await r.json()
+                    if "error" in res:
+                        detail = f"External: {res['error'].get('message', '')}"
+            except Exception:
+                detail = f"External: {e}"
+
+            raise HTTPException(
+                status_code=getattr(r, "status", 500),
+                detail=detail if detail else "Open WebUI: Server Connection Error",
+            )
+
+    elif request.app.state.config.TTS_ENGINE == "azure":
+        try:
+            payload = json.loads(body.decode("utf-8"))
+        except Exception as e:
+            log.exception(e)
+            raise HTTPException(status_code=400, detail="Invalid JSON payload")
+
+        region = request.app.state.config.TTS_AZURE_SPEECH_REGION
+        language = request.app.state.config.TTS_VOICE
+        locale = "-".join(request.app.state.config.TTS_VOICE.split("-")[:1])
+        output_format = request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT
+
+        try:
+            data = f"""<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{locale}">
+                <voice name="{language}">{payload["input"]}</voice>
+            </speak>"""
+            async with aiohttp.ClientSession() as session:
+                async with session.post(
+                    f"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1",
+                    headers={
+                        "Ocp-Apim-Subscription-Key": request.app.state.config.TTS_API_KEY,
+                        "Content-Type": "application/ssml+xml",
+                        "X-Microsoft-OutputFormat": output_format,
+                    },
+                    data=data,
+                ) as r:
+                    r.raise_for_status()
+
+                    async with aiofiles.open(file_path, "wb") as f:
+                        await f.write(await r.read())
+
+                    return FileResponse(file_path)
+
+        except Exception as e:
+            log.exception(e)
+            detail = None
+
+            try:
+                if r.status != 200:
+                    res = await r.json()
+                    if "error" in res:
+                        detail = f"External: {res['error'].get('message', '')}"
+            except Exception:
+                detail = f"External: {e}"
+
+            raise HTTPException(
+                status_code=getattr(r, "status", 500),
+                detail=detail if detail else "Open WebUI: Server Connection Error",
+            )
+
+    elif request.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 = request.app.state.speech_speaker_embeddings_dataset
+
+        speaker_index = 6799
+        try:
+            speaker_index = embeddings_dataset["filename"].index(
+                request.app.state.config.TTS_MODEL
+            )
+        except Exception:
+            pass
+
+        speaker_embedding = torch.tensor(
+            embeddings_dataset[speaker_index]["xvector"]
+        ).unsqueeze(0)
+
+        speech = request.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(request: Request, file_path):
+    print("transcribe", file_path)
+    filename = os.path.basename(file_path)
+    file_dir = os.path.dirname(file_path)
+    id = filename.split(".")[0]
+
+    if request.app.state.config.STT_ENGINE == "":
+        if request.app.state.faster_whisper_model is None:
+            request.app.state.faster_whisper_model = set_faster_whisper_model(
+                request.app.state.config.WHISPER_MODEL
+            )
+
+        model = request.app.state.faster_whisper_model
+        segments, info = model.transcribe(file_path, beam_size=5)
+        log.info(
+            "Detected language '%s' with probability %f"
+            % (info.language, info.language_probability)
+        )
+
+        transcript = "".join([segment.text for segment in list(segments)])
+        data = {"text": transcript.strip()}
+
+        # save the transcript to a json file
+        transcript_file = f"{file_dir}/{id}.json"
+        with open(transcript_file, "w") as f:
+            json.dump(data, f)
+
+        log.debug(data)
+        return data
+    elif request.app.state.config.STT_ENGINE == "openai":
+        if is_mp4_audio(file_path):
+            os.rename(file_path, file_path.replace(".wav", ".mp4"))
+            # Convert MP4 audio file to WAV format
+            convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path)
+
+        r = None
+        try:
+            r = requests.post(
+                url=f"{request.app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions",
+                headers={
+                    "Authorization": f"Bearer {request.app.state.config.STT_OPENAI_API_KEY}"
+                },
+                files={"file": (filename, open(file_path, "rb"))},
+                data={"model": request.app.state.config.STT_MODEL},
+            )
+
+            r.raise_for_status()
+            data = r.json()
+
+            # save the transcript to a json file
+            transcript_file = f"{file_dir}/{id}.json"
+            with open(transcript_file, "w") as f:
+                json.dump(data, f)
+
+            return data
+        except Exception as e:
+            log.exception(e)
+
+            detail = None
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "error" in res:
+                        detail = f"External: {res['error'].get('message', '')}"
+                except Exception:
+                    detail = f"External: {e}"
+
+            raise Exception(detail if detail else "Open WebUI: Server Connection Error")
+
+
+def compress_audio(file_path):
+    if os.path.getsize(file_path) > MAX_FILE_SIZE:
+        file_dir = os.path.dirname(file_path)
+        audio = AudioSegment.from_file(file_path)
+        audio = audio.set_frame_rate(16000).set_channels(1)  # Compress audio
+        compressed_path = f"{file_dir}/{id}_compressed.opus"
+        audio.export(compressed_path, format="opus", bitrate="32k")
+        log.debug(f"Compressed audio to {compressed_path}")
+
+        if (
+            os.path.getsize(compressed_path) > MAX_FILE_SIZE
+        ):  # Still larger than MAX_FILE_SIZE after compression
+            raise Exception(ERROR_MESSAGES.FILE_TOO_LARGE(size=f"{MAX_FILE_SIZE_MB}MB"))
+        return compressed_path
+    else:
+        return file_path
+
+
+@router.post("/transcriptions")
+def transcription(
+    request: Request,
+    file: UploadFile = File(...),
+    user=Depends(get_verified_user),
+):
+    log.info(f"file.content_type: {file.content_type}")
+
+    if file.content_type not in ["audio/mpeg", "audio/wav", "audio/ogg", "audio/x-m4a"]:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED,
+        )
+
+    try:
+        ext = file.filename.split(".")[-1]
+        id = uuid.uuid4()
+
+        filename = f"{id}.{ext}"
+        contents = file.file.read()
+
+        file_dir = f"{CACHE_DIR}/audio/transcriptions"
+        os.makedirs(file_dir, exist_ok=True)
+        file_path = f"{file_dir}/{filename}"
+
+        with open(file_path, "wb") as f:
+            f.write(contents)
+
+        try:
+            try:
+                file_path = compress_audio(file_path)
+            except Exception as e:
+                log.exception(e)
+
+                raise HTTPException(
+                    status_code=status.HTTP_400_BAD_REQUEST,
+                    detail=ERROR_MESSAGES.DEFAULT(e),
+                )
+
+            data = transcribe(request, file_path)
+            file_path = file_path.split("/")[-1]
+            return {**data, "filename": file_path}
+        except Exception as e:
+            log.exception(e)
+
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=ERROR_MESSAGES.DEFAULT(e),
+            )
+
+    except Exception as e:
+        log.exception(e)
+
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+def get_available_models(request: Request) -> list[dict]:
+    available_models = []
+    if request.app.state.config.TTS_ENGINE == "openai":
+        available_models = [{"id": "tts-1"}, {"id": "tts-1-hd"}]
+    elif request.app.state.config.TTS_ENGINE == "elevenlabs":
+        try:
+            response = requests.get(
+                "https://api.elevenlabs.io/v1/models",
+                headers={
+                    "xi-api-key": request.app.state.config.TTS_API_KEY,
+                    "Content-Type": "application/json",
+                },
+                timeout=5,
+            )
+            response.raise_for_status()
+            models = response.json()
+
+            available_models = [
+                {"name": model["name"], "id": model["model_id"]} for model in models
+            ]
+        except requests.RequestException as e:
+            log.error(f"Error fetching voices: {str(e)}")
+    return available_models
+
+
+@router.get("/models")
+async def get_models(request: Request, user=Depends(get_verified_user)):
+    return {"models": get_available_models(request)}
+
+
+def get_available_voices(request) -> dict:
+    """Returns {voice_id: voice_name} dict"""
+    available_voices = {}
+    if request.app.state.config.TTS_ENGINE == "openai":
+        available_voices = {
+            "alloy": "alloy",
+            "echo": "echo",
+            "fable": "fable",
+            "onyx": "onyx",
+            "nova": "nova",
+            "shimmer": "shimmer",
+        }
+    elif request.app.state.config.TTS_ENGINE == "elevenlabs":
+        try:
+            available_voices = get_elevenlabs_voices(
+                api_key=request.app.state.config.TTS_API_KEY
+            )
+        except Exception:
+            # Avoided @lru_cache with exception
+            pass
+    elif request.app.state.config.TTS_ENGINE == "azure":
+        try:
+            region = request.app.state.config.TTS_AZURE_SPEECH_REGION
+            url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/voices/list"
+            headers = {
+                "Ocp-Apim-Subscription-Key": request.app.state.config.TTS_API_KEY
+            }
+
+            response = requests.get(url, headers=headers)
+            response.raise_for_status()
+            voices = response.json()
+
+            for voice in voices:
+                available_voices[voice["ShortName"]] = (
+                    f"{voice['DisplayName']} ({voice['ShortName']})"
+                )
+        except requests.RequestException as e:
+            log.error(f"Error fetching voices: {str(e)}")
+
+    return available_voices
+
+
+@lru_cache
+def get_elevenlabs_voices(api_key: str) -> dict:
+    """
+    Note, set the following in your .env file to use Elevenlabs:
+    AUDIO_TTS_ENGINE=elevenlabs
+    AUDIO_TTS_API_KEY=sk_...  # Your Elevenlabs API key
+    AUDIO_TTS_VOICE=EXAVITQu4vr4xnSDxMaL  # From https://api.elevenlabs.io/v1/voices
+    AUDIO_TTS_MODEL=eleven_multilingual_v2
+    """
+
+    try:
+        # TODO: Add retries
+        response = requests.get(
+            "https://api.elevenlabs.io/v1/voices",
+            headers={
+                "xi-api-key": api_key,
+                "Content-Type": "application/json",
+            },
+        )
+        response.raise_for_status()
+        voices_data = response.json()
+
+        voices = {}
+        for voice in voices_data.get("voices", []):
+            voices[voice["voice_id"]] = voice["name"]
+    except requests.RequestException as e:
+        # Avoid @lru_cache with exception
+        log.error(f"Error fetching voices: {str(e)}")
+        raise RuntimeError(f"Error fetching voices: {str(e)}")
+
+    return voices
+
+
+@router.get("/voices")
+async def get_voices(request: Request, user=Depends(get_verified_user)):
+    return {
+        "voices": [
+            {"id": k, "name": v} for k, v in get_available_voices(request).items()
+        ]
+    }

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

@@ -3,8 +3,9 @@ import uuid
 import time
 import datetime
 import logging
+from aiohttp import ClientSession
 
-from open_webui.apps.webui.models.auths import (
+from open_webui.models.auths import (
     AddUserForm,
     ApiKey,
     Auths,
@@ -17,7 +18,7 @@ from open_webui.apps.webui.models.auths import (
     UpdateProfileForm,
     UserResponse,
 )
-from open_webui.apps.webui.models.users import Users
+from open_webui.models.users import Users
 
 from open_webui.constants import ERROR_MESSAGES, WEBHOOK_MESSAGES
 from open_webui.env import (
@@ -29,10 +30,14 @@ from open_webui.env import (
     SRC_LOG_LEVELS,
 )
 from fastapi import APIRouter, Depends, HTTPException, Request, status
-from fastapi.responses import Response
+from fastapi.responses import RedirectResponse, Response
+from open_webui.config import (
+    OPENID_PROVIDER_URL,
+    ENABLE_OAUTH_SIGNUP,
+)
 from pydantic import BaseModel
 from open_webui.utils.misc import parse_duration, validate_email_format
-from open_webui.utils.utils import (
+from open_webui.utils.auth import (
     create_api_key,
     create_token,
     get_admin_user,
@@ -498,8 +503,31 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
 
 
 @router.get("/signout")
-async def signout(response: Response):
+async def signout(request: Request, response: Response):
     response.delete_cookie("token")
+
+    if ENABLE_OAUTH_SIGNUP.value:
+        oauth_id_token = request.cookies.get("oauth_id_token")
+        if oauth_id_token:
+            try:
+                async with ClientSession() as session:
+                    async with session.get(OPENID_PROVIDER_URL.value) as resp:
+                        if resp.status == 200:
+                            openid_data = await resp.json()
+                            logout_url = openid_data.get("end_session_endpoint")
+                            if logout_url:
+                                response.delete_cookie("oauth_id_token")
+                                return RedirectResponse(
+                                    url=f"{logout_url}?id_token_hint={oauth_id_token}"
+                                )
+                        else:
+                            raise HTTPException(
+                                status_code=resp.status,
+                                detail="Failed to fetch OpenID configuration",
+                            )
+            except Exception as e:
+                raise HTTPException(status_code=500, detail=str(e))
+
     return {"status": True}
 
 

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

@@ -2,15 +2,15 @@ import json
 import logging
 from typing import Optional
 
-from open_webui.apps.webui.models.chats import (
+from open_webui.models.chats import (
     ChatForm,
     ChatImportForm,
     ChatResponse,
     Chats,
     ChatTitleIdResponse,
 )
-from open_webui.apps.webui.models.tags import TagModel, Tags
-from open_webui.apps.webui.models.folders import Folders
+from open_webui.models.tags import TagModel, Tags
+from open_webui.models.folders import Folders
 
 from open_webui.config import ENABLE_ADMIN_CHAT_ACCESS, ENABLE_ADMIN_EXPORT
 from open_webui.constants import ERROR_MESSAGES
@@ -19,7 +19,7 @@ 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.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_permission
 
 log = logging.getLogger(__name__)
@@ -607,7 +607,6 @@ async def add_tag_by_id_and_tag_name(
                 detail=ERROR_MESSAGES.DEFAULT("Tag name cannot be 'None'"),
             )
 
-        print(tags, tag_id)
         if tag_id not in tags:
             Chats.add_chat_tag_by_id_and_user_id_and_tag_name(
                 id, user.id, form_data.name

backend/open_webui/apps/webui/routers/configs.py → backend/open_webui/routers/configs.py

@@ -3,7 +3,7 @@ from pydantic import BaseModel
 
 from typing import Optional
 
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 from open_webui.config import get_config, save_config
 from open_webui.config import BannerModel
 

backend/open_webui/apps/webui/routers/evaluations.py → backend/open_webui/routers/evaluations.py

@@ -2,8 +2,8 @@ from typing import Optional
 from fastapi import APIRouter, Depends, HTTPException, status, Request
 from pydantic import BaseModel
 
-from open_webui.apps.webui.models.users import Users, UserModel
-from open_webui.apps.webui.models.feedbacks import (
+from open_webui.models.users import Users, UserModel
+from open_webui.models.feedbacks import (
     FeedbackModel,
     FeedbackResponse,
     FeedbackForm,
@@ -11,7 +11,7 @@ from open_webui.apps.webui.models.feedbacks import (
 )
 
 from open_webui.constants import ERROR_MESSAGES
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 
 router = APIRouter()
 

backend/open_webui/apps/webui/routers/files.py → backend/open_webui/routers/files.py

@@ -5,27 +5,28 @@ from pathlib import Path
 from typing import Optional
 from pydantic import BaseModel
 import mimetypes
+from urllib.parse import quote
 
 from open_webui.storage.provider import Storage
 
-from open_webui.apps.webui.models.files import (
+from open_webui.models.files import (
     FileForm,
     FileModel,
     FileModelResponse,
     Files,
 )
-from open_webui.apps.retrieval.main import process_file, ProcessFileForm
+from open_webui.routers.retrieval import process_file, ProcessFileForm
 
 from open_webui.config import UPLOAD_DIR
 from open_webui.env import SRC_LOG_LEVELS
 from open_webui.constants import ERROR_MESSAGES
 
 
-from fastapi import APIRouter, Depends, File, HTTPException, UploadFile, status
+from fastapi import APIRouter, Depends, File, HTTPException, UploadFile, status, Request
 from fastapi.responses import FileResponse, StreamingResponse
 
 
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -39,7 +40,9 @@ router = APIRouter()
 
 
 @router.post("/", response_model=FileModelResponse)
-def upload_file(file: UploadFile = File(...), user=Depends(get_verified_user)):
+def upload_file(
+    request: Request, file: UploadFile = File(...), user=Depends(get_verified_user)
+):
     log.info(f"file.content_type: {file.content_type}")
     try:
         unsanitized_filename = file.filename
@@ -68,7 +71,7 @@ def upload_file(file: UploadFile = File(...), user=Depends(get_verified_user)):
         )
 
         try:
-            process_file(ProcessFileForm(file_id=id))
+            process_file(request, ProcessFileForm(file_id=id))
             file_item = Files.get_file_by_id(id=id)
         except Exception as e:
             log.exception(e)
@@ -183,13 +186,15 @@ class ContentForm(BaseModel):
 
 @router.post("/{id}/data/content/update")
 async def update_file_data_content_by_id(
-    id: str, form_data: ContentForm, user=Depends(get_verified_user)
+    request: Request, id: str, form_data: ContentForm, user=Depends(get_verified_user)
 ):
     file = Files.get_file_by_id(id)
 
     if file and (file.user_id == user.id or user.role == "admin"):
         try:
-            process_file(ProcessFileForm(file_id=id, content=form_data.content))
+            process_file(
+                request, ProcessFileForm(file_id=id, content=form_data.content)
+            )
             file = Files.get_file_by_id(id=id)
         except Exception as e:
             log.exception(e)
@@ -218,11 +223,15 @@ async def get_file_content_by_id(id: str, user=Depends(get_verified_user)):
 
             # Check if the file already exists in the cache
             if file_path.is_file():
-                print(f"file_path: {file_path}")
+                # Handle Unicode filenames
+                filename = file.meta.get("name", file.filename)
+                encoded_filename = quote(filename)  # RFC5987 encoding
                 headers = {
-                    "Content-Disposition": f'attachment; filename="{file.meta.get("name", file.filename)}"'
+                    "Content-Disposition": f"attachment; filename*=UTF-8''{encoded_filename}"
                 }
+
                 return FileResponse(file_path, headers=headers)
+
             else:
                 raise HTTPException(
                     status_code=status.HTTP_404_NOT_FOUND,
@@ -279,16 +288,20 @@ async def get_file_content_by_id(id: str, user=Depends(get_verified_user)):
 
     if file and (file.user_id == user.id or user.role == "admin"):
         file_path = file.path
+
+        # Handle Unicode filenames
+        filename = file.meta.get("name", file.filename)
+        encoded_filename = quote(filename)  # RFC5987 encoding
+        headers = {
+            "Content-Disposition": f"attachment; filename*=UTF-8''{encoded_filename}"
+        }
+
         if file_path:
             file_path = Storage.get_file(file_path)
             file_path = Path(file_path)
 
             # Check if the file already exists in the cache
             if file_path.is_file():
-                print(f"file_path: {file_path}")
-                headers = {
-                    "Content-Disposition": f'attachment; filename="{file.meta.get("name", file.filename)}"'
-                }
                 return FileResponse(file_path, headers=headers)
             else:
                 raise HTTPException(
@@ -307,7 +320,7 @@ async def get_file_content_by_id(id: str, user=Depends(get_verified_user)):
             return StreamingResponse(
                 generator(),
                 media_type="text/plain",
-                headers={"Content-Disposition": f"attachment; filename={file_name}"},
+                headers=headers,
             )
     else:
         raise HTTPException(

backend/open_webui/apps/webui/routers/folders.py → backend/open_webui/routers/folders.py

@@ -8,12 +8,12 @@ from pydantic import BaseModel
 import mimetypes
 
 
-from open_webui.apps.webui.models.folders import (
+from open_webui.models.folders import (
     FolderForm,
     FolderModel,
     Folders,
 )
-from open_webui.apps.webui.models.chats import Chats
+from open_webui.models.chats import Chats
 
 from open_webui.config import UPLOAD_DIR
 from open_webui.env import SRC_LOG_LEVELS
@@ -24,7 +24,7 @@ from fastapi import APIRouter, Depends, File, HTTPException, UploadFile, status
 from fastapi.responses import FileResponse, StreamingResponse
 
 
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])

backend/open_webui/apps/webui/routers/functions.py → backend/open_webui/routers/functions.py

@@ -2,17 +2,17 @@ import os
 from pathlib import Path
 from typing import Optional
 
-from open_webui.apps.webui.models.functions import (
+from open_webui.models.functions import (
     FunctionForm,
     FunctionModel,
     FunctionResponse,
     Functions,
 )
-from open_webui.apps.webui.utils import load_function_module_by_id, replace_imports
+from open_webui.utils.plugin import load_function_module_by_id, replace_imports
 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
+from open_webui.utils.auth import get_admin_user, get_verified_user
 
 router = APIRouter()
 

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

@@ -2,7 +2,7 @@ import os
 from pathlib import Path
 from typing import Optional
 
-from open_webui.apps.webui.models.groups import (
+from open_webui.models.groups import (
     Groups,
     GroupForm,
     GroupUpdateForm,
@@ -12,7 +12,7 @@ from open_webui.apps.webui.models.groups import (
 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
+from open_webui.utils.auth import get_admin_user, get_verified_user
 
 router = APIRouter()
 

backend/open_webui/apps/images/main.py → backend/open_webui/routers/images.py

@@ -9,38 +9,24 @@ from pathlib import Path
 from typing import Optional
 
 import requests
-from open_webui.apps.images.utils.comfyui import (
+
+
+from fastapi import Depends, FastAPI, HTTPException, Request, APIRouter
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+
+
+from open_webui.config import CACHE_DIR
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import ENV, SRC_LOG_LEVELS, ENABLE_FORWARD_USER_INFO_HEADERS
+
+from open_webui.utils.auth import get_admin_user, get_verified_user
+from open_webui.utils.images.comfyui import (
     ComfyUIGenerateImageForm,
     ComfyUIWorkflow,
     comfyui_generate_image,
 )
-from open_webui.config import (
-    AUTOMATIC1111_API_AUTH,
-    AUTOMATIC1111_BASE_URL,
-    AUTOMATIC1111_CFG_SCALE,
-    AUTOMATIC1111_SAMPLER,
-    AUTOMATIC1111_SCHEDULER,
-    CACHE_DIR,
-    COMFYUI_BASE_URL,
-    COMFYUI_WORKFLOW,
-    COMFYUI_WORKFLOW_NODES,
-    CORS_ALLOW_ORIGIN,
-    ENABLE_IMAGE_GENERATION,
-    IMAGE_GENERATION_ENGINE,
-    IMAGE_GENERATION_MODEL,
-    IMAGE_SIZE,
-    IMAGE_STEPS,
-    IMAGES_OPENAI_API_BASE_URL,
-    IMAGES_OPENAI_API_KEY,
-    AppConfig,
-)
-from open_webui.constants import ERROR_MESSAGES
-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
-from open_webui.utils.utils import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["IMAGES"])
@@ -48,63 +34,30 @@ 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(
-    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,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-app.state.config = AppConfig()
 
-app.state.config.ENGINE = IMAGE_GENERATION_ENGINE
-app.state.config.ENABLED = ENABLE_IMAGE_GENERATION
+router = APIRouter()
 
-app.state.config.OPENAI_API_BASE_URL = IMAGES_OPENAI_API_BASE_URL
-app.state.config.OPENAI_API_KEY = IMAGES_OPENAI_API_KEY
 
-app.state.config.MODEL = IMAGE_GENERATION_MODEL
-
-app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
-app.state.config.AUTOMATIC1111_API_AUTH = AUTOMATIC1111_API_AUTH
-app.state.config.AUTOMATIC1111_CFG_SCALE = AUTOMATIC1111_CFG_SCALE
-app.state.config.AUTOMATIC1111_SAMPLER = AUTOMATIC1111_SAMPLER
-app.state.config.AUTOMATIC1111_SCHEDULER = AUTOMATIC1111_SCHEDULER
-app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL
-app.state.config.COMFYUI_WORKFLOW = COMFYUI_WORKFLOW
-app.state.config.COMFYUI_WORKFLOW_NODES = COMFYUI_WORKFLOW_NODES
-
-app.state.config.IMAGE_SIZE = IMAGE_SIZE
-app.state.config.IMAGE_STEPS = IMAGE_STEPS
-
-
-@app.get("/config")
+@router.get("/config")
 async def get_config(request: Request, user=Depends(get_admin_user)):
     return {
-        "enabled": app.state.config.ENABLED,
-        "engine": app.state.config.ENGINE,
+        "enabled": request.app.state.config.ENABLE_IMAGE_GENERATION,
+        "engine": request.app.state.config.IMAGE_GENERATION_ENGINE,
         "openai": {
-            "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+            "OPENAI_API_BASE_URL": request.app.state.config.IMAGES_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.IMAGES_OPENAI_API_KEY,
         },
         "automatic1111": {
-            "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
-            "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
-            "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE,
-            "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER,
-            "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER,
+            "AUTOMATIC1111_BASE_URL": request.app.state.config.AUTOMATIC1111_BASE_URL,
+            "AUTOMATIC1111_API_AUTH": request.app.state.config.AUTOMATIC1111_API_AUTH,
+            "AUTOMATIC1111_CFG_SCALE": request.app.state.config.AUTOMATIC1111_CFG_SCALE,
+            "AUTOMATIC1111_SAMPLER": request.app.state.config.AUTOMATIC1111_SAMPLER,
+            "AUTOMATIC1111_SCHEDULER": request.app.state.config.AUTOMATIC1111_SCHEDULER,
         },
         "comfyui": {
-            "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
-            "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW,
-            "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES,
+            "COMFYUI_BASE_URL": request.app.state.config.COMFYUI_BASE_URL,
+            "COMFYUI_WORKFLOW": request.app.state.config.COMFYUI_WORKFLOW,
+            "COMFYUI_WORKFLOW_NODES": request.app.state.config.COMFYUI_WORKFLOW_NODES,
         },
     }
 
@@ -117,7 +70,7 @@ class OpenAIConfigForm(BaseModel):
 class Automatic1111ConfigForm(BaseModel):
     AUTOMATIC1111_BASE_URL: str
     AUTOMATIC1111_API_AUTH: str
-    AUTOMATIC1111_CFG_SCALE: Optional[str]
+    AUTOMATIC1111_CFG_SCALE: Optional[str | float | int]
     AUTOMATIC1111_SAMPLER: Optional[str]
     AUTOMATIC1111_SCHEDULER: Optional[str]
 
@@ -136,133 +89,156 @@ class ConfigForm(BaseModel):
     comfyui: ComfyUIConfigForm
 
 
-@app.post("/config/update")
-async def update_config(form_data: ConfigForm, user=Depends(get_admin_user)):
-    app.state.config.ENGINE = form_data.engine
-    app.state.config.ENABLED = form_data.enabled
+@router.post("/config/update")
+async def update_config(
+    request: Request, form_data: ConfigForm, user=Depends(get_admin_user)
+):
+    request.app.state.config.IMAGE_GENERATION_ENGINE = form_data.engine
+    request.app.state.config.ENABLE_IMAGE_GENERATION = form_data.enabled
 
-    app.state.config.OPENAI_API_BASE_URL = form_data.openai.OPENAI_API_BASE_URL
-    app.state.config.OPENAI_API_KEY = form_data.openai.OPENAI_API_KEY
+    request.app.state.config.IMAGES_OPENAI_API_BASE_URL = (
+        form_data.openai.OPENAI_API_BASE_URL
+    )
+    request.app.state.config.IMAGES_OPENAI_API_KEY = form_data.openai.OPENAI_API_KEY
 
-    app.state.config.AUTOMATIC1111_BASE_URL = (
+    request.app.state.config.AUTOMATIC1111_BASE_URL = (
         form_data.automatic1111.AUTOMATIC1111_BASE_URL
     )
-    app.state.config.AUTOMATIC1111_API_AUTH = (
+    request.app.state.config.AUTOMATIC1111_API_AUTH = (
         form_data.automatic1111.AUTOMATIC1111_API_AUTH
     )
 
-    app.state.config.AUTOMATIC1111_CFG_SCALE = (
+    request.app.state.config.AUTOMATIC1111_CFG_SCALE = (
         float(form_data.automatic1111.AUTOMATIC1111_CFG_SCALE)
         if form_data.automatic1111.AUTOMATIC1111_CFG_SCALE
         else None
     )
-    app.state.config.AUTOMATIC1111_SAMPLER = (
+    request.app.state.config.AUTOMATIC1111_SAMPLER = (
         form_data.automatic1111.AUTOMATIC1111_SAMPLER
         if form_data.automatic1111.AUTOMATIC1111_SAMPLER
         else None
     )
-    app.state.config.AUTOMATIC1111_SCHEDULER = (
+    request.app.state.config.AUTOMATIC1111_SCHEDULER = (
         form_data.automatic1111.AUTOMATIC1111_SCHEDULER
         if form_data.automatic1111.AUTOMATIC1111_SCHEDULER
         else None
     )
 
-    app.state.config.COMFYUI_BASE_URL = form_data.comfyui.COMFYUI_BASE_URL.strip("/")
-    app.state.config.COMFYUI_WORKFLOW = form_data.comfyui.COMFYUI_WORKFLOW
-    app.state.config.COMFYUI_WORKFLOW_NODES = form_data.comfyui.COMFYUI_WORKFLOW_NODES
+    request.app.state.config.COMFYUI_BASE_URL = (
+        form_data.comfyui.COMFYUI_BASE_URL.strip("/")
+    )
+    request.app.state.config.COMFYUI_WORKFLOW = form_data.comfyui.COMFYUI_WORKFLOW
+    request.app.state.config.COMFYUI_WORKFLOW_NODES = (
+        form_data.comfyui.COMFYUI_WORKFLOW_NODES
+    )
 
     return {
-        "enabled": app.state.config.ENABLED,
-        "engine": app.state.config.ENGINE,
+        "enabled": request.app.state.config.ENABLE_IMAGE_GENERATION,
+        "engine": request.app.state.config.IMAGE_GENERATION_ENGINE,
         "openai": {
-            "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+            "OPENAI_API_BASE_URL": request.app.state.config.IMAGES_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": request.app.state.config.IMAGES_OPENAI_API_KEY,
         },
         "automatic1111": {
-            "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
-            "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
-            "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE,
-            "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER,
-            "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER,
+            "AUTOMATIC1111_BASE_URL": request.app.state.config.AUTOMATIC1111_BASE_URL,
+            "AUTOMATIC1111_API_AUTH": request.app.state.config.AUTOMATIC1111_API_AUTH,
+            "AUTOMATIC1111_CFG_SCALE": request.app.state.config.AUTOMATIC1111_CFG_SCALE,
+            "AUTOMATIC1111_SAMPLER": request.app.state.config.AUTOMATIC1111_SAMPLER,
+            "AUTOMATIC1111_SCHEDULER": request.app.state.config.AUTOMATIC1111_SCHEDULER,
         },
         "comfyui": {
-            "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
-            "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW,
-            "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES,
+            "COMFYUI_BASE_URL": request.app.state.config.COMFYUI_BASE_URL,
+            "COMFYUI_WORKFLOW": request.app.state.config.COMFYUI_WORKFLOW,
+            "COMFYUI_WORKFLOW_NODES": request.app.state.config.COMFYUI_WORKFLOW_NODES,
         },
     }
 
 
-def get_automatic1111_api_auth():
-    if app.state.config.AUTOMATIC1111_API_AUTH is None:
+def get_automatic1111_api_auth(request: Request):
+    if request.app.state.config.AUTOMATIC1111_API_AUTH is None:
         return ""
     else:
-        auth1111_byte_string = app.state.config.AUTOMATIC1111_API_AUTH.encode("utf-8")
+        auth1111_byte_string = request.app.state.config.AUTOMATIC1111_API_AUTH.encode(
+            "utf-8"
+        )
         auth1111_base64_encoded_bytes = base64.b64encode(auth1111_byte_string)
         auth1111_base64_encoded_string = auth1111_base64_encoded_bytes.decode("utf-8")
         return f"Basic {auth1111_base64_encoded_string}"
 
 
-@app.get("/config/url/verify")
-async def verify_url(user=Depends(get_admin_user)):
-    if app.state.config.ENGINE == "automatic1111":
+@router.get("/config/url/verify")
+async def verify_url(request: Request, user=Depends(get_admin_user)):
+    if request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111":
         try:
             r = requests.get(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
-                headers={"authorization": get_automatic1111_api_auth()},
+                url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                headers={"authorization": get_automatic1111_api_auth(request)},
             )
             r.raise_for_status()
             return True
         except Exception:
-            app.state.config.ENABLED = False
+            request.app.state.config.ENABLE_IMAGE_GENERATION = False
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL)
-    elif app.state.config.ENGINE == "comfyui":
+    elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui":
         try:
-            r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info")
+            r = requests.get(
+                url=f"{request.app.state.config.COMFYUI_BASE_URL}/object_info"
+            )
             r.raise_for_status()
             return True
         except Exception:
-            app.state.config.ENABLED = False
+            request.app.state.config.ENABLE_IMAGE_GENERATION = False
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL)
     else:
         return True
 
 
-def set_image_model(model: str):
+def set_image_model(request: Request, model: str):
     log.info(f"Setting image model to {model}")
-    app.state.config.MODEL = model
-    if app.state.config.ENGINE in ["", "automatic1111"]:
+    request.app.state.config.IMAGE_GENERATION_MODEL = model
+    if request.app.state.config.IMAGE_GENERATION_ENGINE in ["", "automatic1111"]:
         api_auth = get_automatic1111_api_auth()
         r = requests.get(
-            url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+            url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
             headers={"authorization": api_auth},
         )
         options = r.json()
         if model != options["sd_model_checkpoint"]:
             options["sd_model_checkpoint"] = model
             r = requests.post(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
                 json=options,
                 headers={"authorization": api_auth},
             )
-    return app.state.config.MODEL
+    return request.app.state.config.IMAGE_GENERATION_MODEL
 
 
-def get_image_model():
-    if app.state.config.ENGINE == "openai":
-        return app.state.config.MODEL if app.state.config.MODEL else "dall-e-2"
-    elif app.state.config.ENGINE == "comfyui":
-        return app.state.config.MODEL if app.state.config.MODEL else ""
-    elif app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == "":
+def get_image_model(request):
+    if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai":
+        return (
+            request.app.state.config.IMAGE_GENERATION_MODEL
+            if request.app.state.config.IMAGE_GENERATION_MODEL
+            else "dall-e-2"
+        )
+    elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui":
+        return (
+            request.app.state.config.IMAGE_GENERATION_MODEL
+            if request.app.state.config.IMAGE_GENERATION_MODEL
+            else ""
+        )
+    elif (
+        request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111"
+        or request.app.state.config.IMAGE_GENERATION_ENGINE == ""
+    ):
         try:
             r = requests.get(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
                 headers={"authorization": get_automatic1111_api_auth()},
             )
             options = r.json()
             return options["sd_model_checkpoint"]
         except Exception as e:
-            app.state.config.ENABLED = False
+            request.app.state.config.ENABLE_IMAGE_GENERATION = False
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
 
@@ -272,23 +248,25 @@ class ImageConfigForm(BaseModel):
     IMAGE_STEPS: int
 
 
-@app.get("/image/config")
-async def get_image_config(user=Depends(get_admin_user)):
+@router.get("/image/config")
+async def get_image_config(request: Request, user=Depends(get_admin_user)):
     return {
-        "MODEL": app.state.config.MODEL,
-        "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
-        "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
+        "MODEL": request.app.state.config.IMAGE_GENERATION_MODEL,
+        "IMAGE_SIZE": request.app.state.config.IMAGE_SIZE,
+        "IMAGE_STEPS": request.app.state.config.IMAGE_STEPS,
     }
 
 
-@app.post("/image/config/update")
-async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin_user)):
+@router.post("/image/config/update")
+async def update_image_config(
+    request: Request, form_data: ImageConfigForm, user=Depends(get_admin_user)
+):
 
-    set_image_model(form_data.MODEL)
+    set_image_model(request, form_data.MODEL)
 
     pattern = r"^\d+x\d+$"
     if re.match(pattern, form_data.IMAGE_SIZE):
-        app.state.config.IMAGE_SIZE = form_data.IMAGE_SIZE
+        request.app.state.config.IMAGE_SIZE = form_data.IMAGE_SIZE
     else:
         raise HTTPException(
             status_code=400,
@@ -296,7 +274,7 @@ async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin
         )
 
     if form_data.IMAGE_STEPS >= 0:
-        app.state.config.IMAGE_STEPS = form_data.IMAGE_STEPS
+        request.app.state.config.IMAGE_STEPS = form_data.IMAGE_STEPS
     else:
         raise HTTPException(
             status_code=400,
@@ -304,29 +282,31 @@ async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin
         )
 
     return {
-        "MODEL": app.state.config.MODEL,
-        "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
-        "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
+        "MODEL": request.app.state.config.IMAGE_GENERATION_MODEL,
+        "IMAGE_SIZE": request.app.state.config.IMAGE_SIZE,
+        "IMAGE_STEPS": request.app.state.config.IMAGE_STEPS,
     }
 
 
-@app.get("/models")
-def get_models(user=Depends(get_verified_user)):
+@router.get("/models")
+def get_models(request: Request, user=Depends(get_verified_user)):
     try:
-        if app.state.config.ENGINE == "openai":
+        if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai":
             return [
                 {"id": "dall-e-2", "name": "DALL·E 2"},
                 {"id": "dall-e-3", "name": "DALL·E 3"},
             ]
-        elif app.state.config.ENGINE == "comfyui":
+        elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui":
             # TODO - get models from comfyui
-            r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info")
+            r = requests.get(
+                url=f"{request.app.state.config.COMFYUI_BASE_URL}/object_info"
+            )
             info = r.json()
 
-            workflow = json.loads(app.state.config.COMFYUI_WORKFLOW)
+            workflow = json.loads(request.app.state.config.COMFYUI_WORKFLOW)
             model_node_id = None
 
-            for node in app.state.config.COMFYUI_WORKFLOW_NODES:
+            for node in request.app.state.config.COMFYUI_WORKFLOW_NODES:
                 if node["type"] == "model":
                     if node["node_ids"]:
                         model_node_id = node["node_ids"][0]
@@ -362,10 +342,11 @@ def get_models(user=Depends(get_verified_user)):
                     )
                 )
         elif (
-            app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == ""
+            request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111"
+            or request.app.state.config.IMAGE_GENERATION_ENGINE == ""
         ):
             r = requests.get(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models",
+                url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models",
                 headers={"authorization": get_automatic1111_api_auth()},
             )
             models = r.json()
@@ -376,7 +357,7 @@ def get_models(user=Depends(get_verified_user)):
                 )
             )
     except Exception as e:
-        app.state.config.ENABLED = False
+        request.app.state.config.ENABLE_IMAGE_GENERATION = False
         raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
 
@@ -448,18 +429,21 @@ def save_url_image(url):
         return None
 
 
-@app.post("/generations")
+@router.post("/generations")
 async def image_generations(
+    request: Request,
     form_data: GenerateImageForm,
     user=Depends(get_verified_user),
 ):
-    width, height = tuple(map(int, app.state.config.IMAGE_SIZE.split("x")))
+    width, height = tuple(map(int, request.app.state.config.IMAGE_SIZE.split("x")))
 
     r = None
     try:
-        if app.state.config.ENGINE == "openai":
+        if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai":
             headers = {}
-            headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}"
+            headers["Authorization"] = (
+                f"Bearer {request.app.state.config.IMAGES_OPENAI_API_KEY}"
+            )
             headers["Content-Type"] = "application/json"
 
             if ENABLE_FORWARD_USER_INFO_HEADERS:
@@ -470,14 +454,16 @@ async def image_generations(
 
             data = {
                 "model": (
-                    app.state.config.MODEL
-                    if app.state.config.MODEL != ""
+                    request.app.state.config.IMAGE_GENERATION_MODEL
+                    if request.app.state.config.IMAGE_GENERATION_MODEL != ""
                     else "dall-e-2"
                 ),
                 "prompt": form_data.prompt,
                 "n": form_data.n,
                 "size": (
-                    form_data.size if form_data.size else app.state.config.IMAGE_SIZE
+                    form_data.size
+                    if form_data.size
+                    else request.app.state.config.IMAGE_SIZE
                 ),
                 "response_format": "b64_json",
             }
@@ -485,7 +471,7 @@ async def image_generations(
             # Use asyncio.to_thread for the requests.post call
             r = await asyncio.to_thread(
                 requests.post,
-                url=f"{app.state.config.OPENAI_API_BASE_URL}/images/generations",
+                url=f"{request.app.state.config.IMAGES_OPENAI_API_BASE_URL}/images/generations",
                 json=data,
                 headers=headers,
             )
@@ -505,7 +491,7 @@ async def image_generations(
 
             return images
 
-        elif app.state.config.ENGINE == "comfyui":
+        elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui":
             data = {
                 "prompt": form_data.prompt,
                 "width": width,
@@ -513,8 +499,8 @@ async def image_generations(
                 "n": form_data.n,
             }
 
-            if app.state.config.IMAGE_STEPS is not None:
-                data["steps"] = app.state.config.IMAGE_STEPS
+            if request.app.state.config.IMAGE_STEPS is not None:
+                data["steps"] = request.app.state.config.IMAGE_STEPS
 
             if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
@@ -523,18 +509,18 @@ async def image_generations(
                 **{
                     "workflow": ComfyUIWorkflow(
                         **{
-                            "workflow": app.state.config.COMFYUI_WORKFLOW,
-                            "nodes": app.state.config.COMFYUI_WORKFLOW_NODES,
+                            "workflow": request.app.state.config.COMFYUI_WORKFLOW,
+                            "nodes": request.app.state.config.COMFYUI_WORKFLOW_NODES,
                         }
                     ),
                     **data,
                 }
             )
             res = await comfyui_generate_image(
-                app.state.config.MODEL,
+                request.app.state.config.IMAGE_GENERATION_MODEL,
                 form_data,
                 user.id,
-                app.state.config.COMFYUI_BASE_URL,
+                request.app.state.config.COMFYUI_BASE_URL,
             )
             log.debug(f"res: {res}")
 
@@ -551,7 +537,8 @@ async def image_generations(
             log.debug(f"images: {images}")
             return images
         elif (
-            app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == ""
+            request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111"
+            or request.app.state.config.IMAGE_GENERATION_ENGINE == ""
         ):
             if form_data.model:
                 set_image_model(form_data.model)
@@ -563,25 +550,25 @@ async def image_generations(
                 "height": height,
             }
 
-            if app.state.config.IMAGE_STEPS is not None:
-                data["steps"] = app.state.config.IMAGE_STEPS
+            if request.app.state.config.IMAGE_STEPS is not None:
+                data["steps"] = request.app.state.config.IMAGE_STEPS
 
             if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
 
-            if app.state.config.AUTOMATIC1111_CFG_SCALE:
-                data["cfg_scale"] = app.state.config.AUTOMATIC1111_CFG_SCALE
+            if request.app.state.config.AUTOMATIC1111_CFG_SCALE:
+                data["cfg_scale"] = request.app.state.config.AUTOMATIC1111_CFG_SCALE
 
-            if app.state.config.AUTOMATIC1111_SAMPLER:
-                data["sampler_name"] = app.state.config.AUTOMATIC1111_SAMPLER
+            if request.app.state.config.AUTOMATIC1111_SAMPLER:
+                data["sampler_name"] = request.app.state.config.AUTOMATIC1111_SAMPLER
 
-            if app.state.config.AUTOMATIC1111_SCHEDULER:
-                data["scheduler"] = app.state.config.AUTOMATIC1111_SCHEDULER
+            if request.app.state.config.AUTOMATIC1111_SCHEDULER:
+                data["scheduler"] = request.app.state.config.AUTOMATIC1111_SCHEDULER
 
             # Use asyncio.to_thread for the requests.post call
             r = await asyncio.to_thread(
                 requests.post,
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img",
+                url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img",
                 json=data,
                 headers={"authorization": get_automatic1111_api_auth()},
             )

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

@@ -4,19 +4,19 @@ from pydantic import BaseModel
 from fastapi import APIRouter, Depends, HTTPException, status, Request
 import logging
 
-from open_webui.apps.webui.models.knowledge import (
+from open_webui.models.knowledge import (
     Knowledges,
     KnowledgeForm,
     KnowledgeResponse,
     KnowledgeUserResponse,
 )
-from open_webui.apps.webui.models.files import Files, FileModel
-from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
-from open_webui.apps.retrieval.main import process_file, ProcessFileForm
+from open_webui.models.files import Files, FileModel
+from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT
+from open_webui.routers.retrieval 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.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_access, has_permission
 
 
@@ -242,6 +242,7 @@ class KnowledgeFileIdForm(BaseModel):
 
 @router.post("/{id}/file/add", response_model=Optional[KnowledgeFilesResponse])
 def add_file_to_knowledge_by_id(
+    request: Request,
     id: str,
     form_data: KnowledgeFileIdForm,
     user=Depends(get_verified_user),
@@ -274,7 +275,9 @@ def add_file_to_knowledge_by_id(
 
     # Add content to the vector database
     try:
-        process_file(ProcessFileForm(file_id=form_data.file_id, collection_name=id))
+        process_file(
+            request, ProcessFileForm(file_id=form_data.file_id, collection_name=id)
+        )
     except Exception as e:
         log.debug(e)
         raise HTTPException(
@@ -318,6 +321,7 @@ def add_file_to_knowledge_by_id(
 
 @router.post("/{id}/file/update", response_model=Optional[KnowledgeFilesResponse])
 def update_file_from_knowledge_by_id(
+    request: Request,
     id: str,
     form_data: KnowledgeFileIdForm,
     user=Depends(get_verified_user),
@@ -349,7 +353,9 @@ def update_file_from_knowledge_by_id(
 
     # Add content to the vector database
     try:
-        process_file(ProcessFileForm(file_id=form_data.file_id, collection_name=id))
+        process_file(
+            request, ProcessFileForm(file_id=form_data.file_id, collection_name=id)
+        )
     except Exception as e:
         raise HTTPException(
             status_code=status.HTTP_400_BAD_REQUEST,

backend/open_webui/apps/webui/routers/memories.py → backend/open_webui/routers/memories.py

@@ -3,9 +3,9 @@ from pydantic import BaseModel
 import logging
 from typing import Optional
 
-from open_webui.apps.webui.models.memories import Memories, MemoryModel
-from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT
-from open_webui.utils.utils import get_verified_user
+from open_webui.models.memories import Memories, MemoryModel
+from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT
+from open_webui.utils.auth import get_verified_user
 from open_webui.env import SRC_LOG_LEVELS
 
 

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

@@ -1,6 +1,6 @@
 from typing import Optional
 
-from open_webui.apps.webui.models.models import (
+from open_webui.models.models import (
     ModelForm,
     ModelModel,
     ModelResponse,
@@ -11,7 +11,7 @@ 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.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_access, has_permission
 
 

backend/open_webui/apps/openai/main.py → backend/open_webui/routers/openai.py

@@ -10,15 +10,15 @@ from aiocache import cached
 import requests
 
 
-from open_webui.apps.webui.models.models import Models
+from fastapi import Depends, FastAPI, HTTPException, Request, APIRouter
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import FileResponse, StreamingResponse
+from pydantic import BaseModel
+from starlette.background import BackgroundTask
+
+from open_webui.models.models import Models
 from open_webui.config import (
     CACHE_DIR,
-    CORS_ALLOW_ORIGIN,
-    ENABLE_OPENAI_API,
-    OPENAI_API_BASE_URLS,
-    OPENAI_API_KEYS,
-    OPENAI_API_CONFIGS,
-    AppConfig,
 )
 from open_webui.env import (
     AIOHTTP_CLIENT_TIMEOUT,
@@ -29,18 +29,14 @@ from open_webui.env import (
 
 from open_webui.constants import ERROR_MESSAGES
 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
-from pydantic import BaseModel
-from starlette.background import BackgroundTask
+
 
 from open_webui.utils.payload import (
     apply_model_params_to_body_openai,
     apply_model_system_prompt_to_body,
 )
 
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_access
 
 
@@ -48,36 +44,69 @@ log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["OPENAI"])
 
 
-app = FastAPI(
-    docs_url="/docs" if ENV == "dev" else None,
-    openapi_url="/openapi.json" if ENV == "dev" else None,
-    redoc_url=None,
-)
+##########################################
+#
+# Utility functions
+#
+##########################################
 
 
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=CORS_ALLOW_ORIGIN,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
+async def send_get_request(url, key=None):
+    timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
+    try:
+        async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session:
+            async with session.get(
+                url, headers={**({"Authorization": f"Bearer {key}"} if key else {})}
+            ) as response:
+                return await response.json()
+    except Exception as e:
+        # Handle connection error here
+        log.error(f"Connection error: {e}")
+        return None
+
+
+async def cleanup_response(
+    response: Optional[aiohttp.ClientResponse],
+    session: Optional[aiohttp.ClientSession],
+):
+    if response:
+        response.close()
+    if session:
+        await session.close()
+
 
-app.state.config = AppConfig()
+def openai_o1_handler(payload):
+    """
+    Handle O1 specific parameters
+    """
+    if "max_tokens" in payload:
+        # Remove "max_tokens" from the payload
+        payload["max_completion_tokens"] = payload["max_tokens"]
+        del payload["max_tokens"]
 
-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.config.OPENAI_API_CONFIGS = OPENAI_API_CONFIGS
+    # Fix: O1 does not support the "system" parameter, Modify "system" to "user"
+    if payload["messages"][0]["role"] == "system":
+        payload["messages"][0]["role"] = "user"
 
+    return payload
 
-@app.get("/config")
-async def get_config(user=Depends(get_admin_user)):
+
+##########################################
+#
+# API routes
+#
+##########################################
+
+router = APIRouter()
+
+
+@router.get("/config")
+async def get_config(request: Request, user=Depends(get_admin_user)):
     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,
+        "ENABLE_OPENAI_API": request.app.state.config.ENABLE_OPENAI_API,
+        "OPENAI_API_BASE_URLS": request.app.state.config.OPENAI_API_BASE_URLS,
+        "OPENAI_API_KEYS": request.app.state.config.OPENAI_API_KEYS,
+        "OPENAI_API_CONFIGS": request.app.state.config.OPENAI_API_CONFIGS,
     }
 
 
@@ -88,50 +117,56 @@ class OpenAIConfigForm(BaseModel):
     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
-
-    app.state.config.OPENAI_API_BASE_URLS = form_data.OPENAI_API_BASE_URLS
-    app.state.config.OPENAI_API_KEYS = form_data.OPENAI_API_KEYS
+@router.post("/config/update")
+async def update_config(
+    request: Request, form_data: OpenAIConfigForm, user=Depends(get_admin_user)
+):
+    request.app.state.config.ENABLE_OPENAI_API = form_data.ENABLE_OPENAI_API
+    request.app.state.config.OPENAI_API_BASE_URLS = form_data.OPENAI_API_BASE_URLS
+    request.app.state.config.OPENAI_API_KEYS = form_data.OPENAI_API_KEYS
 
     # 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(request.app.state.config.OPENAI_API_KEYS) != len(
+        request.app.state.config.OPENAI_API_BASE_URLS
     ):
-        if len(app.state.config.OPENAI_API_KEYS) > len(
-            app.state.config.OPENAI_API_BASE_URLS
+        if len(request.app.state.config.OPENAI_API_KEYS) > len(
+            request.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)
-            ]
+            request.app.state.config.OPENAI_API_KEYS = (
+                request.app.state.config.OPENAI_API_KEYS[
+                    : len(request.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)
+            request.app.state.config.OPENAI_API_KEYS += [""] * (
+                len(request.app.state.config.OPENAI_API_BASE_URLS)
+                - len(request.app.state.config.OPENAI_API_KEYS)
             )
 
-    app.state.config.OPENAI_API_CONFIGS = form_data.OPENAI_API_CONFIGS
+    request.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):
+    config_urls = request.app.state.config.OPENAI_API_CONFIGS.keys()
+    for idx, url in enumerate(request.app.state.config.OPENAI_API_BASE_URLS):
         if url not in config_urls:
-            app.state.config.OPENAI_API_CONFIGS.pop(url, None)
+            request.app.state.config.OPENAI_API_CONFIGS.pop(url, None)
 
     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,
+        "ENABLE_OPENAI_API": request.app.state.config.ENABLE_OPENAI_API,
+        "OPENAI_API_BASE_URLS": request.app.state.config.OPENAI_API_BASE_URLS,
+        "OPENAI_API_KEYS": request.app.state.config.OPENAI_API_KEYS,
+        "OPENAI_API_CONFIGS": request.app.state.config.OPENAI_API_CONFIGS,
     }
 
 
-@app.post("/audio/speech")
+@router.post("/audio/speech")
 async def speech(request: Request, user=Depends(get_verified_user)):
     idx = None
     try:
-        idx = app.state.config.OPENAI_API_BASE_URLS.index("https://api.openai.com/v1")
+        idx = request.app.state.config.OPENAI_API_BASE_URLS.index(
+            "https://api.openai.com/v1"
+        )
+
         body = await request.body()
         name = hashlib.sha256(body).hexdigest()
 
@@ -144,23 +179,35 @@ async def speech(request: Request, user=Depends(get_verified_user)):
         if file_path.is_file():
             return FileResponse(file_path)
 
-        headers = {}
-        headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEYS[idx]}"
-        headers["Content-Type"] = "application/json"
-        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
+        url = request.app.state.config.OPENAI_API_BASE_URLS[idx]
+
         r = None
         try:
             r = requests.post(
-                url=f"{app.state.config.OPENAI_API_BASE_URLS[idx]}/audio/speech",
+                url=f"{url}/audio/speech",
                 data=body,
-                headers=headers,
+                headers={
+                    "Content-Type": "application/json",
+                    "Authorization": f"Bearer {request.app.state.config.OPENAI_API_KEYS[idx]}",
+                    **(
+                        {
+                            "HTTP-Referer": "https://openwebui.com/",
+                            "X-Title": "Open WebUI",
+                        }
+                        if "openrouter.ai" in url
+                        else {}
+                    ),
+                    **(
+                        {
+                            "X-OpenWebUI-User-Name": user.name,
+                            "X-OpenWebUI-User-Id": user.id,
+                            "X-OpenWebUI-User-Email": user.email,
+                            "X-OpenWebUI-User-Role": user.role,
+                        }
+                        if ENABLE_FORWARD_USER_INFO_HEADERS
+                        else {}
+                    ),
+                },
                 stream=True,
             )
 
@@ -179,115 +226,62 @@ async def speech(request: Request, user=Depends(get_verified_user)):
 
         except Exception as e:
             log.exception(e)
-            error_detail = "Open WebUI: Server Connection Error"
+
+            detail = None
             if r is not None:
                 try:
                     res = r.json()
                     if "error" in res:
-                        error_detail = f"External: {res['error']}"
+                        detail = f"External: {res['error']}"
                 except Exception:
-                    error_detail = f"External: {e}"
+                    detail = f"External: {e}"
 
             raise HTTPException(
-                status_code=r.status_code if r else 500, detail=error_detail
+                status_code=r.status_code if r else 500,
+                detail=detail if detail else "Open WebUI: Server Connection Error",
             )
 
     except ValueError:
         raise HTTPException(status_code=401, detail=ERROR_MESSAGES.OPENAI_NOT_FOUND)
 
 
-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, headers=headers) as response:
-                return await response.json()
-    except Exception as e:
-        # Handle connection error here
-        log.error(f"Connection error: {e}")
-        return None
-
-
-async def cleanup_response(
-    response: Optional[aiohttp.ClientResponse],
-    session: Optional[aiohttp.ClientSession],
-):
-    if response:
-        response.close()
-    if session:
-        await session.close()
-
-
-def merge_models_lists(model_lists):
-    log.debug(f"merge_models_lists {model_lists}")
-    merged_list = []
-
-    for idx, models in enumerate(model_lists):
-        if models is not None and "error" not in models:
-            merged_list.extend(
-                [
-                    {
-                        **model,
-                        "name": model.get("name", model["id"]),
-                        "owned_by": "openai",
-                        "openai": model,
-                        "urlIdx": idx,
-                    }
-                    for model in models
-                    if "api.openai.com"
-                    not in app.state.config.OPENAI_API_BASE_URLS[idx]
-                    or not any(
-                        name in model["id"]
-                        for name in [
-                            "babbage",
-                            "dall-e",
-                            "davinci",
-                            "embedding",
-                            "tts",
-                            "whisper",
-                        ]
-                    )
-                ]
-            )
-
-    return merged_list
-
-
-async def get_all_models_responses() -> list:
-    if not app.state.config.ENABLE_OPENAI_API:
+async def get_all_models_responses(request: Request) -> list:
+    if not request.app.state.config.ENABLE_OPENAI_API:
         return []
 
     # Check if API KEYS length is same than API URLS length
-    num_urls = len(app.state.config.OPENAI_API_BASE_URLS)
-    num_keys = len(app.state.config.OPENAI_API_KEYS)
+    num_urls = len(request.app.state.config.OPENAI_API_BASE_URLS)
+    num_keys = len(request.app.state.config.OPENAI_API_KEYS)
 
     if num_keys != num_urls:
         # if there are more keys than urls, remove the extra keys
         if num_keys > num_urls:
-            new_keys = app.state.config.OPENAI_API_KEYS[:num_urls]
-            app.state.config.OPENAI_API_KEYS = new_keys
+            new_keys = request.app.state.config.OPENAI_API_KEYS[:num_urls]
+            request.app.state.config.OPENAI_API_KEYS = new_keys
         # if there are more urls than keys, add empty keys
         else:
-            app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys)
-
-    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])
+            request.app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys)
+
+    request_tasks = []
+    for idx, url in enumerate(request.app.state.config.OPENAI_API_BASE_URLS):
+        if url not in request.app.state.config.OPENAI_API_CONFIGS:
+            request_tasks.append(
+                send_get_request(
+                    f"{url}/models", request.app.state.config.OPENAI_API_KEYS[idx]
+                )
             )
         else:
-            api_config = app.state.config.OPENAI_API_CONFIGS.get(url, {})
+            api_config = request.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]
+                    request_tasks.append(
+                        send_get_request(
+                            f"{url}/models",
+                            request.app.state.config.OPENAI_API_KEYS[idx],
                         )
                     )
                 else:
@@ -305,16 +299,18 @@ async def get_all_models_responses() -> list:
                         ],
                     }
 
-                    tasks.append(asyncio.ensure_future(asyncio.sleep(0, model_list)))
+                    request_tasks.append(
+                        asyncio.ensure_future(asyncio.sleep(0, model_list))
+                    )
             else:
-                tasks.append(asyncio.ensure_future(asyncio.sleep(0, None)))
+                request_tasks.append(asyncio.ensure_future(asyncio.sleep(0, None)))
 
-    responses = await asyncio.gather(*tasks)
+    responses = await asyncio.gather(*request_tasks)
 
     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, {})
+            url = request.app.state.config.OPENAI_API_BASE_URLS[idx]
+            api_config = request.app.state.config.OPENAI_API_CONFIGS.get(url, {})
 
             prefix_id = api_config.get("prefix_id", None)
 
@@ -325,18 +321,30 @@ async def get_all_models_responses() -> list:
                     model["id"] = f"{prefix_id}.{model['id']}"
 
     log.debug(f"get_all_models:responses() {responses}")
-
     return responses
 
 
+async def get_filtered_models(models, 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)
+    return filtered_models
+
+
 @cached(ttl=3)
-async def get_all_models() -> dict[str, list]:
+async def get_all_models(request: Request) -> dict[str, list]:
     log.info("get_all_models()")
 
-    if not app.state.config.ENABLE_OPENAI_API:
+    if not request.app.state.config.ENABLE_OPENAI_API:
         return {"data": []}
 
-    responses = await get_all_models_responses()
+    responses = await get_all_models_responses(request)
 
     def extract_data(response):
         if response and "data" in response:
@@ -345,41 +353,86 @@ async def get_all_models() -> dict[str, list]:
             return response
         return None
 
+    def merge_models_lists(model_lists):
+        log.debug(f"merge_models_lists {model_lists}")
+        merged_list = []
+
+        for idx, models in enumerate(model_lists):
+            if models is not None and "error" not in models:
+                merged_list.extend(
+                    [
+                        {
+                            **model,
+                            "name": model.get("name", model["id"]),
+                            "owned_by": "openai",
+                            "openai": model,
+                            "urlIdx": idx,
+                        }
+                        for model in models
+                        if "api.openai.com"
+                        not in request.app.state.config.OPENAI_API_BASE_URLS[idx]
+                        or not any(
+                            name in model["id"]
+                            for name in [
+                                "babbage",
+                                "dall-e",
+                                "davinci",
+                                "embedding",
+                                "tts",
+                                "whisper",
+                            ]
+                        )
+                    ]
+                )
+
+        return merged_list
+
     models = {"data": merge_models_lists(map(extract_data, responses))}
     log.debug(f"models: {models}")
 
+    request.app.state.OPENAI_MODELS = {model["id"]: model for model in models["data"]}
     return models
 
 
-@app.get("/models")
-@app.get("/models/{url_idx}")
-async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_user)):
+@router.get("/models")
+@router.get("/models/{url_idx}")
+async def get_models(
+    request: Request, url_idx: Optional[int] = None, user=Depends(get_verified_user)
+):
     models = {
         "data": [],
     }
 
     if url_idx is None:
-        models = await get_all_models()
+        models = await get_all_models(request)
     else:
-        url = app.state.config.OPENAI_API_BASE_URLS[url_idx]
-        key = app.state.config.OPENAI_API_KEYS[url_idx]
-
-        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
+        url = request.app.state.config.OPENAI_API_BASE_URLS[url_idx]
+        key = request.app.state.config.OPENAI_API_KEYS[url_idx]
 
         r = None
-
-        timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
-        async with aiohttp.ClientSession(timeout=timeout) as session:
+        async with aiohttp.ClientSession(
+            timeout=aiohttp.ClientTimeout(
+                total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST
+            )
+        ) as session:
             try:
-                async with session.get(f"{url}/models", headers=headers) as r:
+                async with session.get(
+                    f"{url}/models",
+                    headers={
+                        "Authorization": f"Bearer {key}",
+                        "Content-Type": "application/json",
+                        **(
+                            {
+                                "X-OpenWebUI-User-Name": user.name,
+                                "X-OpenWebUI-User-Id": user.id,
+                                "X-OpenWebUI-User-Email": user.email,
+                                "X-OpenWebUI-User-Role": user.role,
+                            }
+                            if ENABLE_FORWARD_USER_INFO_HEADERS
+                            else {}
+                        ),
+                    },
+                ) as r:
                     if r.status != 200:
                         # Extract response error details if available
                         error_detail = f"HTTP Error: {r.status}"
@@ -413,27 +466,16 @@ async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_us
             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" and not BYPASS_MODEL_ACCESS_CONTROL:
-        # 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
+        models["data"] = get_filtered_models(models, user)
 
     return models
 
@@ -443,21 +485,24 @@ class ConnectionVerificationForm(BaseModel):
     key: str
 
 
-@app.post("/verify")
+@router.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:
+    async with aiohttp.ClientSession(
+        timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST)
+    ) as session:
         try:
-            async with session.get(f"{url}/models", headers=headers) as r:
+            async with session.get(
+                f"{url}/models",
+                headers={
+                    "Authorization": f"Bearer {key}",
+                    "Content-Type": "application/json",
+                },
+            ) as r:
                 if r.status != 200:
                     # Extract response error details if available
                     error_detail = f"HTTP Error: {r.status}"
@@ -472,26 +517,24 @@ async def verify_connection(
         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.post("/chat/completions")
+@router.post("/chat/completions")
 async def generate_chat_completion(
+    request: Request,
     form_data: dict,
     user=Depends(get_verified_user),
     bypass_filter: Optional[bool] = False,
 ):
     idx = 0
     payload = {**form_data}
-
     if "metadata" in payload:
         del payload["metadata"]
 
@@ -526,15 +569,7 @@ async def generate_chat_completion(
                 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,
-    )
-
+    model = request.app.state.OPENAI_MODELS.get(model_id)
     if model:
         idx = model["urlIdx"]
     else:
@@ -544,11 +579,11 @@ async def generate_chat_completion(
         )
 
     # Get the API config for the model
-    api_config = app.state.config.OPENAI_API_CONFIGS.get(
-        app.state.config.OPENAI_API_BASE_URLS[idx], {}
+    api_config = request.app.state.config.OPENAI_API_CONFIGS.get(
+        request.app.state.config.OPENAI_API_BASE_URLS[idx], {}
     )
-    prefix_id = api_config.get("prefix_id", None)
 
+    prefix_id = api_config.get("prefix_id", None)
     if prefix_id:
         payload["model"] = payload["model"].replace(f"{prefix_id}.", "")
 
@@ -561,43 +596,26 @@ async def generate_chat_completion(
             "role": user.role,
         }
 
-    url = app.state.config.OPENAI_API_BASE_URLS[idx]
-    key = app.state.config.OPENAI_API_KEYS[idx]
+    url = request.app.state.config.OPENAI_API_BASE_URLS[idx]
+    key = request.app.state.config.OPENAI_API_KEYS[idx]
 
     # 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:
-            # Remove "max_completion_tokens" from the payload
-            payload["max_tokens"] = payload["max_completion_tokens"]
-            del payload["max_completion_tokens"]
-    else:
-        if is_o1 and "max_tokens" in payload:
+    if is_o1:
+        payload = openai_o1_handler(payload)
+    elif "api.openai.com" not in url:
+        # Remove "max_tokens" from the payload for backward compatibility
+        if "max_tokens" in payload:
             payload["max_completion_tokens"] = payload["max_tokens"]
             del payload["max_tokens"]
-        if "max_tokens" in payload and "max_completion_tokens" in payload:
-            del payload["max_tokens"]
 
-    # Fix: O1 does not support the "system" parameter, Modify "system" to "user"
-    if is_o1 and payload["messages"][0]["role"] == "system":
-        payload["messages"][0]["role"] = "user"
+    # TODO: check if below is needed
+    # if "max_tokens" in payload and "max_completion_tokens" in payload:
+    #     del payload["max_tokens"]
 
     # Convert the modified body back to JSON
     payload = json.dumps(payload)
 
-    headers = {}
-    headers["Authorization"] = f"Bearer {key}"
-    headers["Content-Type"] = "application/json"
-    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
     streaming = False
@@ -607,11 +625,33 @@ async def generate_chat_completion(
         session = aiohttp.ClientSession(
             trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT)
         )
+
         r = await session.request(
             method="POST",
             url=f"{url}/chat/completions",
             data=payload,
-            headers=headers,
+            headers={
+                "Authorization": f"Bearer {key}",
+                "Content-Type": "application/json",
+                **(
+                    {
+                        "HTTP-Referer": "https://openwebui.com/",
+                        "X-Title": "Open WebUI",
+                    }
+                    if "openrouter.ai" in url
+                    else {}
+                ),
+                **(
+                    {
+                        "X-OpenWebUI-User-Name": user.name,
+                        "X-OpenWebUI-User-Id": user.id,
+                        "X-OpenWebUI-User-Email": user.email,
+                        "X-OpenWebUI-User-Role": user.role,
+                    }
+                    if ENABLE_FORWARD_USER_INFO_HEADERS
+                    else {}
+                ),
+            },
         )
 
         # Check if response is SSE
@@ -636,14 +676,18 @@ async def generate_chat_completion(
             return response
     except Exception as e:
         log.exception(e)
-        error_detail = "Open WebUI: Server Connection Error"
+
+        detail = None
         if isinstance(response, dict):
             if "error" in response:
-                error_detail = f"{response['error']['message'] if 'message' in response['error'] else response['error']}"
+                detail = f"{response['error']['message'] if 'message' in response['error'] else response['error']}"
         elif isinstance(response, str):
-            error_detail = response
+            detail = response
 
-        raise HTTPException(status_code=r.status if r else 500, detail=error_detail)
+        raise HTTPException(
+            status_code=r.status if r else 500,
+            detail=detail if detail else "Open WebUI: Server Connection Error",
+        )
     finally:
         if not streaming and session:
             if r:
@@ -651,25 +695,17 @@ async def generate_chat_completion(
             await session.close()
 
 
-@app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
+@router.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
 async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
-    idx = 0
+    """
+    Deprecated: proxy all requests to OpenAI API
+    """
 
     body = await request.body()
 
-    url = app.state.config.OPENAI_API_BASE_URLS[idx]
-    key = app.state.config.OPENAI_API_KEYS[idx]
-
-    target_url = f"{url}/{path}"
-
-    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
+    idx = 0
+    url = request.app.state.config.OPENAI_API_BASE_URLS[idx]
+    key = request.app.state.config.OPENAI_API_KEYS[idx]
 
     r = None
     session = None
@@ -679,11 +715,23 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
         session = aiohttp.ClientSession(trust_env=True)
         r = await session.request(
             method=request.method,
-            url=target_url,
+            url=f"{url}/{path}",
             data=body,
-            headers=headers,
+            headers={
+                "Authorization": f"Bearer {key}",
+                "Content-Type": "application/json",
+                **(
+                    {
+                        "X-OpenWebUI-User-Name": user.name,
+                        "X-OpenWebUI-User-Id": user.id,
+                        "X-OpenWebUI-User-Email": user.email,
+                        "X-OpenWebUI-User-Role": user.role,
+                    }
+                    if ENABLE_FORWARD_USER_INFO_HEADERS
+                    else {}
+                ),
+            },
         )
-
         r.raise_for_status()
 
         # Check if response is SSE
@@ -700,18 +748,23 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
         else:
             response_data = await r.json()
             return response_data
+
     except Exception as e:
         log.exception(e)
-        error_detail = "Open WebUI: Server Connection Error"
+
+        detail = None
         if r is not None:
             try:
                 res = await r.json()
                 print(res)
                 if "error" in res:
-                    error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
+                    detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
             except Exception:
-                error_detail = f"External: {e}"
-        raise HTTPException(status_code=r.status if r else 500, detail=error_detail)
+                detail = f"External: {e}"
+        raise HTTPException(
+            status_code=r.status if r else 500,
+            detail=detail if detail else "Open WebUI: Server Connection Error",
+        )
     finally:
         if not streaming and session:
             if r:

backend/open_webui/routers/pipelines.py

@@ -0,0 +1,496 @@
+from fastapi import (
+    Depends,
+    FastAPI,
+    File,
+    Form,
+    HTTPException,
+    Request,
+    UploadFile,
+    status,
+    APIRouter,
+)
+import os
+import logging
+import shutil
+import requests
+from pydantic import BaseModel
+from starlette.responses import FileResponse
+from typing import Optional
+
+from open_webui.env import SRC_LOG_LEVELS
+from open_webui.config import CACHE_DIR
+from open_webui.constants import ERROR_MESSAGES
+
+
+from open_webui.routers.openai import get_all_models_responses
+
+from open_webui.utils.auth import get_admin_user
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MAIN"])
+
+
+##################################
+#
+# Pipeline Middleware
+#
+##################################
+
+
+def get_sorted_filters(model_id, models):
+    filters = [
+        model
+        for model in models.values()
+        if "pipeline" in model
+        and "type" in model["pipeline"]
+        and model["pipeline"]["type"] == "filter"
+        and (
+            model["pipeline"]["pipelines"] == ["*"]
+            or any(
+                model_id == target_model_id
+                for target_model_id in model["pipeline"]["pipelines"]
+            )
+        )
+    ]
+    sorted_filters = sorted(filters, key=lambda x: x["pipeline"]["priority"])
+    return sorted_filters
+
+
+def process_pipeline_inlet_filter(request, payload, user, models):
+    user = {"id": user.id, "email": user.email, "name": user.name, "role": user.role}
+    model_id = payload["model"]
+
+    sorted_filters = get_sorted_filters(model_id, models)
+    model = models[model_id]
+
+    if "pipeline" in model:
+        sorted_filters.append(model)
+
+    for filter in sorted_filters:
+        r = None
+        try:
+            urlIdx = filter["urlIdx"]
+
+            url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+            key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+            if key == "":
+                continue
+
+            headers = {"Authorization": f"Bearer {key}"}
+            r = requests.post(
+                f"{url}/{filter['id']}/filter/inlet",
+                headers=headers,
+                json={
+                    "user": user,
+                    "body": payload,
+                },
+            )
+
+            r.raise_for_status()
+            payload = r.json()
+        except Exception as e:
+            # Handle connection error here
+            print(f"Connection error: {e}")
+
+            if r is not None:
+                res = r.json()
+                if "detail" in res:
+                    raise Exception(r.status_code, res["detail"])
+
+    return payload
+
+
+def process_pipeline_outlet_filter(request, payload, user, models):
+    user = {"id": user.id, "email": user.email, "name": user.name, "role": user.role}
+    model_id = payload["model"]
+
+    sorted_filters = get_sorted_filters(model_id, models)
+    model = models[model_id]
+
+    if "pipeline" in model:
+        sorted_filters = [model] + sorted_filters
+
+    for filter in sorted_filters:
+        r = None
+        try:
+            urlIdx = filter["urlIdx"]
+
+            url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+            key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+            if key != "":
+                r = requests.post(
+                    f"{url}/{filter['id']}/filter/outlet",
+                    headers={"Authorization": f"Bearer {key}"},
+                    json={
+                        "user": {
+                            "id": user.id,
+                            "name": user.name,
+                            "email": user.email,
+                            "role": user.role,
+                        },
+                        "body": data,
+                    },
+                )
+
+                r.raise_for_status()
+                data = r.json()
+        except Exception as e:
+            # Handle connection error here
+            print(f"Connection error: {e}")
+
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "detail" in res:
+                        return Exception(r.status_code, res)
+                except Exception:
+                    pass
+
+            else:
+                pass
+
+    return payload
+
+
+##################################
+#
+# Pipelines Endpoints
+#
+##################################
+
+router = APIRouter()
+
+
+@router.get("/list")
+async def get_pipelines_list(request: Request, user=Depends(get_admin_user)):
+    responses = await get_all_models_responses(request)
+    log.debug(f"get_pipelines_list: get_openai_models_responses returned {responses}")
+
+    urlIdxs = [
+        idx
+        for idx, response in enumerate(responses)
+        if response is not None and "pipelines" in response
+    ]
+
+    return {
+        "data": [
+            {
+                "url": request.app.state.config.OPENAI_API_BASE_URLS[urlIdx],
+                "idx": urlIdx,
+            }
+            for urlIdx in urlIdxs
+        ]
+    }
+
+
+@router.post("/upload")
+async def upload_pipeline(
+    request: Request,
+    urlIdx: int = Form(...),
+    file: UploadFile = File(...),
+    user=Depends(get_admin_user),
+):
+    print("upload_pipeline", urlIdx, file.filename)
+    # Check if the uploaded file is a python file
+    if not (file.filename and file.filename.endswith(".py")):
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail="Only Python (.py) files are allowed.",
+        )
+
+    upload_folder = f"{CACHE_DIR}/pipelines"
+    os.makedirs(upload_folder, exist_ok=True)
+    file_path = os.path.join(upload_folder, file.filename)
+
+    r = None
+    try:
+        # Save the uploaded file
+        with open(file_path, "wb") as buffer:
+            shutil.copyfileobj(file.file, buffer)
+
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        with open(file_path, "rb") as f:
+            files = {"file": f}
+            r = requests.post(
+                f"{url}/pipelines/upload",
+                headers={"Authorization": f"Bearer {key}"},
+                files=files,
+            )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        status_code = status.HTTP_404_NOT_FOUND
+        if r is not None:
+            status_code = r.status_code
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=status_code,
+            detail=detail if detail else "Pipeline not found",
+        )
+    finally:
+        # Ensure the file is deleted after the upload is completed or on failure
+        if os.path.exists(file_path):
+            os.remove(file_path)
+
+
+class AddPipelineForm(BaseModel):
+    url: str
+    urlIdx: int
+
+
+@router.post("/add")
+async def add_pipeline(
+    request: Request, form_data: AddPipelineForm, user=Depends(get_admin_user)
+):
+    r = None
+    try:
+        urlIdx = form_data.urlIdx
+
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.post(
+            f"{url}/pipelines/add",
+            headers={"Authorization": f"Bearer {key}"},
+            json={"url": form_data.url},
+        )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )
+
+
+class DeletePipelineForm(BaseModel):
+    id: str
+    urlIdx: int
+
+
+@router.delete("/delete")
+async def delete_pipeline(
+    request: Request, form_data: DeletePipelineForm, user=Depends(get_admin_user)
+):
+    r = None
+    try:
+        urlIdx = form_data.urlIdx
+
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.delete(
+            f"{url}/pipelines/delete",
+            headers={"Authorization": f"Bearer {key}"},
+            json={"id": form_data.id},
+        )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )
+
+
+@router.get("/")
+async def get_pipelines(
+    request: Request, urlIdx: Optional[int] = None, user=Depends(get_admin_user)
+):
+    r = None
+    try:
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.get(f"{url}/pipelines", headers={"Authorization": f"Bearer {key}"})
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )
+
+
+@router.get("/{pipeline_id}/valves")
+async def get_pipeline_valves(
+    request: Request,
+    urlIdx: Optional[int],
+    pipeline_id: str,
+    user=Depends(get_admin_user),
+):
+    r = None
+    try:
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.get(
+            f"{url}/{pipeline_id}/valves", headers={"Authorization": f"Bearer {key}"}
+        )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )
+
+
+@router.get("/{pipeline_id}/valves/spec")
+async def get_pipeline_valves_spec(
+    request: Request,
+    urlIdx: Optional[int],
+    pipeline_id: str,
+    user=Depends(get_admin_user),
+):
+    r = None
+    try:
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.get(
+            f"{url}/{pipeline_id}/valves/spec",
+            headers={"Authorization": f"Bearer {key}"},
+        )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )
+
+
+@router.post("/{pipeline_id}/valves/update")
+async def update_pipeline_valves(
+    request: Request,
+    urlIdx: Optional[int],
+    pipeline_id: str,
+    form_data: dict,
+    user=Depends(get_admin_user),
+):
+    r = None
+    try:
+        url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx]
+        key = request.app.state.config.OPENAI_API_KEYS[urlIdx]
+
+        r = requests.post(
+            f"{url}/{pipeline_id}/valves/update",
+            headers={"Authorization": f"Bearer {key}"},
+            json={**form_data},
+        )
+
+        r.raise_for_status()
+        data = r.json()
+
+        return {**data}
+    except Exception as e:
+        # Handle connection error here
+        print(f"Connection error: {e}")
+
+        detail = None
+
+        if r is not None:
+            try:
+                res = r.json()
+                if "detail" in res:
+                    detail = res["detail"]
+            except Exception:
+                pass
+
+        raise HTTPException(
+            status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND),
+            detail=detail if detail else "Pipeline not found",
+        )

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

@@ -1,6 +1,6 @@
 from typing import Optional
 
-from open_webui.apps.webui.models.prompts import (
+from open_webui.models.prompts import (
     PromptForm,
     PromptUserResponse,
     PromptModel,
@@ -8,7 +8,7 @@ from open_webui.apps.webui.models.prompts import (
 )
 from open_webui.constants import ERROR_MESSAGES
 from fastapi import APIRouter, Depends, HTTPException, status, Request
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_access, has_permission
 
 router = APIRouter()

backend/open_webui/routers/tasks.py

@@ -0,0 +1,512 @@
+from fastapi import APIRouter, Depends, HTTPException, Response, status, Request
+from fastapi.responses import JSONResponse, RedirectResponse
+
+from pydantic import BaseModel
+from typing import Optional
+import logging
+
+from open_webui.utils.chat import generate_chat_completion
+from open_webui.utils.task import (
+    title_generation_template,
+    query_generation_template,
+    autocomplete_generation_template,
+    tags_generation_template,
+    emoji_generation_template,
+    moa_response_generation_template,
+)
+from open_webui.utils.auth import get_admin_user, get_verified_user
+from open_webui.constants import TASKS
+
+from open_webui.routers.pipelines import process_pipeline_inlet_filter
+from open_webui.utils.task import get_task_model_id
+
+from open_webui.config import (
+    DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE,
+    DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE,
+    DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE,
+    DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE,
+    DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE,
+    DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE,
+)
+from open_webui.env import SRC_LOG_LEVELS
+
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+router = APIRouter()
+
+
+##################################
+#
+# Task Endpoints
+#
+##################################
+
+
+@router.get("/config")
+async def get_task_config(request: Request, user=Depends(get_verified_user)):
+    return {
+        "TASK_MODEL": request.app.state.config.TASK_MODEL,
+        "TASK_MODEL_EXTERNAL": request.app.state.config.TASK_MODEL_EXTERNAL,
+        "TITLE_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE,
+        "ENABLE_AUTOCOMPLETE_GENERATION": request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION,
+        "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH,
+        "TAGS_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE,
+        "ENABLE_TAGS_GENERATION": request.app.state.config.ENABLE_TAGS_GENERATION,
+        "ENABLE_SEARCH_QUERY_GENERATION": request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION,
+        "ENABLE_RETRIEVAL_QUERY_GENERATION": request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION,
+        "QUERY_GENERATION_PROMPT_TEMPLATE": request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE,
+        "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE,
+    }
+
+
+class TaskConfigForm(BaseModel):
+    TASK_MODEL: Optional[str]
+    TASK_MODEL_EXTERNAL: Optional[str]
+    TITLE_GENERATION_PROMPT_TEMPLATE: str
+    ENABLE_AUTOCOMPLETE_GENERATION: bool
+    AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH: int
+    TAGS_GENERATION_PROMPT_TEMPLATE: str
+    ENABLE_TAGS_GENERATION: bool
+    ENABLE_SEARCH_QUERY_GENERATION: bool
+    ENABLE_RETRIEVAL_QUERY_GENERATION: bool
+    QUERY_GENERATION_PROMPT_TEMPLATE: str
+    TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE: str
+
+
+@router.post("/config/update")
+async def update_task_config(
+    request: Request, form_data: TaskConfigForm, user=Depends(get_admin_user)
+):
+    request.app.state.config.TASK_MODEL = form_data.TASK_MODEL
+    request.app.state.config.TASK_MODEL_EXTERNAL = form_data.TASK_MODEL_EXTERNAL
+    request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE = (
+        form_data.TITLE_GENERATION_PROMPT_TEMPLATE
+    )
+
+    request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION = (
+        form_data.ENABLE_AUTOCOMPLETE_GENERATION
+    )
+    request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH = (
+        form_data.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH
+    )
+
+    request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE = (
+        form_data.TAGS_GENERATION_PROMPT_TEMPLATE
+    )
+    request.app.state.config.ENABLE_TAGS_GENERATION = form_data.ENABLE_TAGS_GENERATION
+    request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION = (
+        form_data.ENABLE_SEARCH_QUERY_GENERATION
+    )
+    request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION = (
+        form_data.ENABLE_RETRIEVAL_QUERY_GENERATION
+    )
+
+    request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE = (
+        form_data.QUERY_GENERATION_PROMPT_TEMPLATE
+    )
+    request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = (
+        form_data.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE
+    )
+
+    return {
+        "TASK_MODEL": request.app.state.config.TASK_MODEL,
+        "TASK_MODEL_EXTERNAL": request.app.state.config.TASK_MODEL_EXTERNAL,
+        "TITLE_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE,
+        "ENABLE_AUTOCOMPLETE_GENERATION": request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION,
+        "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH,
+        "TAGS_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE,
+        "ENABLE_TAGS_GENERATION": request.app.state.config.ENABLE_TAGS_GENERATION,
+        "ENABLE_SEARCH_QUERY_GENERATION": request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION,
+        "ENABLE_RETRIEVAL_QUERY_GENERATION": request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION,
+        "QUERY_GENERATION_PROMPT_TEMPLATE": request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE,
+        "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE,
+    }
+
+
+@router.post("/title/completions")
+async def generate_title(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+    models = request.app.state.MODELS
+
+    model_id = form_data["model"]
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(
+        f"generating chat title using model {task_model_id} for user {user.email} "
+    )
+
+    if request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE != "":
+        template = request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE
+    else:
+        template = DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE
+
+    content = title_generation_template(
+        template,
+        form_data["messages"],
+        {
+            "name": user.name,
+            "location": user.info.get("location") if user.info else None,
+        },
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": False,
+        **(
+            {"max_tokens": 50}
+            if models[task_model_id]["owned_by"] == "ollama"
+            else {
+                "max_completion_tokens": 50,
+            }
+        ),
+        "metadata": {
+            "task": str(TASKS.TITLE_GENERATION),
+            "task_body": form_data,
+            "chat_id": form_data.get("chat_id", None),
+        },
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )
+
+
+@router.post("/tags/completions")
+async def generate_chat_tags(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+
+    if not request.app.state.config.ENABLE_TAGS_GENERATION:
+        return JSONResponse(
+            status_code=status.HTTP_200_OK,
+            content={"detail": "Tags generation is disabled"},
+        )
+
+    models = request.app.state.MODELS
+
+    model_id = form_data["model"]
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(
+        f"generating chat tags using model {task_model_id} for user {user.email} "
+    )
+
+    if request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE != "":
+        template = request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE
+    else:
+        template = DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE
+
+    content = tags_generation_template(
+        template, form_data["messages"], {"name": user.name}
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": False,
+        "metadata": {
+            "task": str(TASKS.TAGS_GENERATION),
+            "task_body": form_data,
+            "chat_id": form_data.get("chat_id", None),
+        },
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )
+
+
+@router.post("/queries/completions")
+async def generate_queries(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+
+    type = form_data.get("type")
+    if type == "web_search":
+        if not request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=f"Search query generation is disabled",
+            )
+    elif type == "retrieval":
+        if not request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=f"Query generation is disabled",
+            )
+
+    models = request.app.state.MODELS
+
+    model_id = form_data["model"]
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(
+        f"generating {type} queries using model {task_model_id} for user {user.email}"
+    )
+
+    if (request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE).strip() != "":
+        template = request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE
+    else:
+        template = DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE
+
+    content = query_generation_template(
+        template, form_data["messages"], {"name": user.name}
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": False,
+        "metadata": {
+            "task": str(TASKS.QUERY_GENERATION),
+            "task_body": form_data,
+            "chat_id": form_data.get("chat_id", None),
+        },
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )
+
+
+@router.post("/auto/completions")
+async def generate_autocompletion(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+    if not request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=f"Autocompletion generation is disabled",
+        )
+
+    type = form_data.get("type")
+    prompt = form_data.get("prompt")
+    messages = form_data.get("messages")
+
+    if request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH > 0:
+        if (
+            len(prompt)
+            > request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH
+        ):
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=f"Input prompt exceeds maximum length of {request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH}",
+            )
+
+    models = request.app.state.MODELS
+
+    model_id = form_data["model"]
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(
+        f"generating autocompletion using model {task_model_id} for user {user.email}"
+    )
+
+    if (request.app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE).strip() != "":
+        template = request.app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE
+    else:
+        template = DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE
+
+    content = autocomplete_generation_template(
+        template, prompt, messages, type, {"name": user.name}
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": False,
+        "metadata": {
+            "task": str(TASKS.AUTOCOMPLETE_GENERATION),
+            "task_body": form_data,
+            "chat_id": form_data.get("chat_id", None),
+        },
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )
+
+
+@router.post("/emoji/completions")
+async def generate_emoji(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+
+    models = request.app.state.MODELS
+
+    model_id = form_data["model"]
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(f"generating emoji using model {task_model_id} for user {user.email} ")
+
+    template = DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE
+
+    content = emoji_generation_template(
+        template,
+        form_data["prompt"],
+        {
+            "name": user.name,
+            "location": user.info.get("location") if user.info else None,
+        },
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": False,
+        **(
+            {"max_tokens": 4}
+            if models[task_model_id]["owned_by"] == "ollama"
+            else {
+                "max_completion_tokens": 4,
+            }
+        ),
+        "chat_id": form_data.get("chat_id", None),
+        "metadata": {"task": str(TASKS.EMOJI_GENERATION), "task_body": form_data},
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )
+
+
+@router.post("/moa/completions")
+async def generate_moa_response(
+    request: Request, form_data: dict, user=Depends(get_verified_user)
+):
+
+    models = request.app.state.MODELS
+    model_id = form_data["model"]
+
+    if model_id not in models:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Model not found",
+        )
+
+    # Check if the user has a custom task model
+    # If the user has a custom task model, use that model
+    task_model_id = get_task_model_id(
+        model_id,
+        request.app.state.config.TASK_MODEL,
+        request.app.state.config.TASK_MODEL_EXTERNAL,
+        models,
+    )
+
+    log.debug(f"generating MOA model {task_model_id} for user {user.email} ")
+
+    template = DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE
+
+    content = moa_response_generation_template(
+        template,
+        form_data["prompt"],
+        form_data["responses"],
+    )
+
+    payload = {
+        "model": task_model_id,
+        "messages": [{"role": "user", "content": content}],
+        "stream": form_data.get("stream", False),
+        "chat_id": form_data.get("chat_id", None),
+        "metadata": {
+            "task": str(TASKS.MOA_RESPONSE_GENERATION),
+            "task_body": form_data,
+        },
+    }
+
+    try:
+        return await generate_chat_completion(request, form_data=payload, user=user)
+    except Exception as e:
+        return JSONResponse(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            content={"detail": str(e)},
+        )

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

@@ -1,19 +1,19 @@
 from pathlib import Path
 from typing import Optional
 
-from open_webui.apps.webui.models.tools import (
+from open_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.utils.plugin import load_tools_module_by_id, replace_imports
 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.tools import get_tools_specs
-from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_verified_user
 from open_webui.utils.access_control import has_access, has_permission
 
 

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

@@ -1,9 +1,9 @@
 import logging
 from typing import Optional
 
-from open_webui.apps.webui.models.auths import Auths
-from open_webui.apps.webui.models.chats import Chats
-from open_webui.apps.webui.models.users import (
+from open_webui.models.auths import Auths
+from open_webui.models.chats import Chats
+from open_webui.models.users import (
     UserModel,
     UserRoleUpdateForm,
     Users,
@@ -14,7 +14,7 @@ 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_password_hash, get_verified_user
+from open_webui.utils.auth import get_admin_user, get_password_hash, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])