matrix-pygmalion-bot/runpod/runpod-worker-transformers/Dockerfile

ARG BASE_IMAGE=nvidia/cuda:11.6.2-cudnn8-devel-ubuntu20.04
FROM ${BASE_IMAGE} as dev-base

WORKDIR /
SHELL ["/bin/bash", "-o", "pipefail", "-c"]
ENV DEBIAN_FRONTEND noninteractive\
    SHELL=/bin/bash

RUN apt-get update --yes && \
    # - apt-get upgrade is run to patch known vulnerabilities in apt-get packages as
    #   the ubuntu base image is rebuilt too seldom sometimes (less than once a month)
    apt-get upgrade --yes && \
    apt install --yes --no-install-recommends \
    build-essential \
    ca-certificates \
    git \
    git-lfs \
    wget \
    curl \
    bash \
    libgl1 \
    software-properties-common \
    openssh-server && \
    apt-get clean && rm -rf /var/lib/apt/lists/* && \
    echo "en_US.UTF-8 UTF-8" > /etc/locale.gen

RUN apt-key del 7fa2af80 && \
    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/3bf863cc.pub

RUN add-apt-repository ppa:deadsnakes/ppa && \
    apt-get install python3.10 python3.10-dev python3.10-venv python3-pip -y --no-install-recommends && \
    apt-get clean && rm -rf /var/lib/apt/lists/*

RUN pip install --upgrade pip && \
    pip install huggingface-hub && \
    pip install diffusers && \
    pip install safetensors && \
    pip install torch torchvision torchaudio --extra-index-url=https://download.pytorch.org/whl/cu116 && \
    pip install bitsandbytes && \
    pip install transformers accelerate xformers triton && \
    pip install runpod

RUN mkdir /workspace
WORKDIR /workspace

COPY model_fetcher.py /workspace/
RUN python model_fetcher.py --model_name=${MODEL_NAME}
#RUN git lfs install && \
#    git clone --depth 1 https://huggingface.co/${MODEL_NAME}

COPY runpod_infer.py /workspace/
COPY test_input.json /workspace/

CMD python -u runpod_infer.py --model_name=${MODEL_NAME}