gpt4all/gpt4all-api/gpt4all_api/Dockerfile.buildkit

# syntax=docker/dockerfile:1.0.0-experimental
FROM tiangolo/uvicorn-gunicorn:python3.11

ARG MODEL_BIN=ggml-mpt-7b-chat.bin

# Put first so anytime this file changes other cached layers are invalidated.
COPY gpt4all_api/requirements.txt /requirements.txt

RUN pip install --upgrade pip

# Run various pip install commands with ssh keys from host machine.
RUN --mount=type=ssh pip install -r /requirements.txt && \
  rm -Rf /root/.cache && rm -Rf /tmp/pip-install*

# Finally, copy app and client.
COPY gpt4all_api/app /app

RUN mkdir -p /models

# Include the following line to bake a model into the image and not have to download it on API start.
RUN wget -q --show-progress=off https://gpt4all.io/models/${MODEL_BIN} -P /models \
  && md5sum /models/${MODEL_BIN}