2023-10-08 22:38:42 +00:00
# syntax=docker/dockerfile:1
2024-03-14 10:18:27 +00:00
######## WebUI frontend ########
FROM node:21-bullseye-slim as build
2023-10-21 23:14:12 +00:00
2023-11-15 00:28:51 +00:00
WORKDIR /app
2024-03-14 10:18:27 +00:00
RUN apt-get update \
&& apt-get install -y --no-install-recommends wget \
# cleanup
&& rm -rf /var/lib/apt/lists/*
2024-01-08 04:50:09 +00:00
# wget embedding model weight from alpine (does not exist from slim-buster)
2024-03-14 10:18:27 +00:00
#RUN wget "https://chroma-onnx-models.s3.amazonaws.com/all-MiniLM-L6-v2/onnx.tar.gz" -O - | \
# tar -xzf - -C /app
2024-01-08 04:50:09 +00:00
2024-01-25 10:08:35 +00:00
COPY package.json package-lock.json ./
2024-01-05 05:15:13 +00:00
RUN npm ci
2024-01-05 04:32:51 +00:00
2024-01-05 05:15:13 +00:00
COPY . .
RUN npm run build
2023-10-08 22:38:42 +00:00
2024-03-14 10:18:27 +00:00
######## WebUI backend ########
2024-01-07 16:28:35 +00:00
FROM python:3.11-slim-bookworm as base
2023-11-15 00:28:51 +00:00
2024-03-14 10:18:27 +00:00
## Basis ##
ENV ENV = prod \
PORT = 8080
2024-01-05 02:55:15 +00:00
2024-03-14 10:18:27 +00:00
## Basis URL Config ##
ENV OLLAMA_BASE_URL = "/ollama" \
OPENAI_API_BASE_URL = ""
2024-01-05 02:55:15 +00:00
2024-03-14 10:18:27 +00:00
## API Key and Security Config ##
ENV OPENAI_API_KEY = "" \
WEBUI_SECRET_KEY = "" \
SCARF_NO_ANALYTICS = true \
DO_NOT_TRACK = true
2023-11-15 00:28:51 +00:00
2024-03-14 10:18:27 +00:00
#### Preloaded models ##########################################################
## whisper TTS Settings ##
ENV WHISPER_MODEL = "base" \
WHISPER_MODEL_DIR = "/app/backend/data/cache/whisper/models"
2024-02-10 02:11:01 +00:00
2024-03-14 10:18:27 +00:00
## RAG Embedding Model Settings ##
2024-02-17 18:38:29 +00:00
# any sentence transformer model; models to use can be found at https://huggingface.co/models?library=sentence-transformers
# Leaderboard: https://huggingface.co/spaces/mteb/leaderboard
2024-02-19 06:51:17 +00:00
# for better persormance and multilangauge support use "intfloat/multilingual-e5-large" (~2.5GB) or "intfloat/multilingual-e5-base" (~1.5GB)
2024-02-17 18:38:29 +00:00
# IMPORTANT: If you change the default model (all-MiniLM-L6-v2) and vice versa, you aren't able to use RAG Chat with your previous documents loaded in the WebUI! You need to re-embed them.
2024-03-14 10:18:27 +00:00
ENV RAG_EMBEDDING_MODEL = "all-MiniLM-L6-v2" \
# device type for whisper tts and embbeding models - "cpu" (default), "cuda" (nvidia gpu and CUDA required) or "mps" (apple silicon) - choosing this right can lead to better performance
RAG_EMBEDDING_MODEL_DEVICE_TYPE = "cpu" \
RAG_EMBEDDING_MODEL_DIR = "/app/backend/data/cache/embedding/models" \
SENTENCE_TRANSFORMERS_HOME = "/app/backend/data/cache/embedding/models"
#### Preloaded models ##########################################################
2024-02-17 18:38:29 +00:00
2023-11-15 00:28:51 +00:00
WORKDIR /app/backend
2024-01-23 12:48:27 +00:00
# install python dependencies
2023-11-15 00:28:51 +00:00
COPY ./backend/requirements.txt ./requirements.txt
2024-01-08 05:22:37 +00:00
2024-03-14 10:18:27 +00:00
RUN pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu --no-cache-dir \
&& pip3 install -r requirements.txt --no-cache-dir
2024-01-08 04:21:32 +00:00
2024-03-14 10:18:27 +00:00
# install required packages
2024-01-23 07:11:50 +00:00
RUN apt-get update \
2024-03-14 10:18:27 +00:00
# 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
2024-01-23 07:11:50 +00:00
&& rm -rf /var/lib/apt/lists/*
2024-02-17 18:38:29 +00:00
# preload embedding model
2024-02-19 06:51:17 +00:00
RUN 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['RAG_EMBEDDING_MODEL_DEVICE_TYPE'])"
2024-02-17 18:38:29 +00:00
# preload tts model
2024-02-19 06:51:17 +00:00
RUN python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='auto', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])"
2024-02-13 14:11:53 +00:00
2024-01-23 12:48:27 +00:00
# copy embedding weight from build
2024-03-14 10:18:27 +00:00
# RUN mkdir -p /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2
# COPY --from=build /app/onnx /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2/onnx
2024-01-23 12:48:27 +00:00
# copy built frontend files
COPY --from= build /app/build /app/build
2024-02-23 08:54:22 +00:00
COPY --from= build /app/CHANGELOG.md /app/CHANGELOG.md
COPY --from= build /app/package.json /app/package.json
2024-01-23 12:48:27 +00:00
# copy backend files
2023-11-15 00:28:51 +00:00
COPY ./backend .
2024-03-07 14:13:51 +00:00
CMD [ "bash" , "start.sh" ]