All in one Dockerfile for including Ollama

Dockerfile

@@ -2,6 +2,7 @@
 # Initialize device type args
 ARG USE_CUDA=false
 ARG USE_MPS=false
+ARG INCLUDE_OLLAMA=false
 
 ######## WebUI frontend ########
 FROM node:21-alpine3.19 as build
@@ -29,10 +30,12 @@ FROM python:3.11-slim-bookworm as base
 # Use args
 ARG USE_CUDA
 ARG USE_MPS
+ARG INCLUDE_OLLAMA
 
 ## Basis ##
 ENV ENV=prod \
-    PORT=8080
+    PORT=8080 \
+    INCLUDE_OLLAMA_ENV=${INCLUDE_OLLAMA}
 
 ## Basis URL Config ##
 ENV OLLAMA_BASE_URL="/ollama" \
@@ -88,14 +91,28 @@ RUN if [ "$USE_CUDA" = "true" ]; then \
         python -c "import os; from chromadb.utils import embedding_functions; sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name=os.environ['RAG_EMBEDDING_MODEL'], device=os.environ['DEVICE_TYPE'])"; \
     fi
 
-#  install required packages
-RUN apt-get update \
-    # Install pandoc and netcat
-    && apt-get install -y --no-install-recommends pandoc netcat-openbsd \
-    # for RAG OCR
-    && apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 \
-    # cleanup
-    && rm -rf /var/lib/apt/lists/*
+
+RUN if [ "$INCLUDE_OLLAMA" = "true" ]; then \
+        apt-get update && \
+        # Install pandoc and netcat
+        apt-get install -y --no-install-recommends pandoc netcat-openbsd && \
+        # for RAG OCR
+        apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
+        # install helper tools
+        apt-get install -y --no-install-recommends curl && \
+        # install ollama
+        curl -fsSL https://ollama.com/install.sh | sh && \
+        # cleanup
+        rm -rf /var/lib/apt/lists/*; \
+    else \
+        apt-get update && \
+        # Install pandoc and netcat
+        apt-get install -y --no-install-recommends pandoc netcat-openbsd && \
+        # for RAG OCR
+        apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
+        # cleanup
+        rm -rf /var/lib/apt/lists/*; \
+    fi
 
 
 

Dockerfile-cuda

@@ -81,9 +81,14 @@ WORKDIR /app/backend
 #    apk del /var/cache/apk/*.tbz2
 
 # Install only the dependencies in the container, python will come from the base image used
-RUN apk update && \
-    apk add --no-install-recommends ffmpeg libsm6 libxext6 pandoc netcat-openbsd && \
-    apk del /var/cache/apk/*.tbz2
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 pandoc netcat-openbsd && \
+    rm -rf /var/cache/apk/*.tbz2
+
+# Install python and pip
+RUN apt-get update && \
+    apt-get install -y python3 python3-pip && \
+    rm -rf /var/cache/apk/*.tbz2
 
 COPY ./backend/requirements.txt ./requirements.txt
 RUN pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 --no-cache-dir && \

Dockerfile-ollama

@@ -64,7 +64,7 @@ RUN pip3 install -r requirements.txt --no-cache-dir
 # Install pandoc and netcat
 # RUN python -c "import pypandoc; pypandoc.download_pandoc()"
 RUN apt-get update \
-    && apt-get install -y pandoc netcat-openbsd \
+    && apt-get install -y pandoc netcat-openbsd curl \
     && rm -rf /var/lib/apt/lists/*
 
 # preload embedding model

backend/config.py

@@ -208,6 +208,7 @@ OLLAMA_API_BASE_URL = os.environ.get(
 )
 
 OLLAMA_BASE_URL = os.environ.get("OLLAMA_BASE_URL", "")
+INCLUDE_OLLAMA = os.environ.get("OLLAMA_BASE_URL", "false")
 
 
 if OLLAMA_BASE_URL == "" and OLLAMA_API_BASE_URL != "":
@@ -219,7 +220,10 @@ if OLLAMA_BASE_URL == "" and OLLAMA_API_BASE_URL != "":
 
 if ENV == "prod":
     if OLLAMA_BASE_URL == "/ollama":
-        OLLAMA_BASE_URL = "http://host.docker.internal:11434"
+        if INCLUDE_OLLAMA:
+            OLLAMA_BASE_URL = "http://localhost:11434"
+        else:    
+            OLLAMA_BASE_URL = "http://host.docker.internal:11434"
 
 
 OLLAMA_BASE_URLS = os.environ.get("OLLAMA_BASE_URLS", "")

backend/start.sh

@@ -1,5 +1,7 @@
 #!/usr/bin/env bash
 
+INCLUDE_OLLAMA=${INCLUDE_OLLAMA_ENV:-false}
+
 SCRIPT_DIR=$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )
 cd "$SCRIPT_DIR" || exit
 
@@ -7,16 +9,21 @@ KEY_FILE=.webui_secret_key
 
 PORT="${PORT:-8080}"
 if test "$WEBUI_SECRET_KEY $WEBUI_JWT_SECRET_KEY" = " "; then
-  echo No WEBUI_SECRET_KEY provided
+  echo "No WEBUI_SECRET_KEY provided"
 
   if ! [ -e "$KEY_FILE" ]; then
-    echo Generating WEBUI_SECRET_KEY
+    echo "Generating WEBUI_SECRET_KEY"
     # Generate a random value to use as a WEBUI_SECRET_KEY in case the user didn't provide one.
-    echo $(head -c 12 /dev/random | base64) > $KEY_FILE
+    echo $(head -c 12 /dev/random | base64) > "$KEY_FILE"
   fi
 
-  echo Loading WEBUI_SECRET_KEY from $KEY_FILE
-  WEBUI_SECRET_KEY=`cat $KEY_FILE`
+  echo "Loading WEBUI_SECRET_KEY from $KEY_FILE"
+  WEBUI_SECRET_KEY=$(cat "$KEY_FILE")
+fi
+
+if [ "$INCLUDE_OLLAMA" = "true" ]; then
+    echo "INCLUDE_OLLAMA is set to true, starting ollama serve."
+    ollama serve &
 fi
 
-WEBUI_SECRET_KEY="$WEBUI_SECRET_KEY" exec uvicorn main:app --host 0.0.0.0 --port "$PORT" --forwarded-allow-ips '*'
+WEBUI_SECRET_KEY="$WEBUI_SECRET_KEY" exec uvicorn main:app --host 0.0.0.0 --port "$PORT" --forwarded-allow-ips '*'