Dockerfile

FROM nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04

ENV DEBIAN_FRONTEND=noninteractive
ENV PYTHONUNBUFFERED=1

RUN apt-get update && apt-get install -y --no-install-recommends \
    python3 python3-pip python3-dev git curl ca-certificates \
 && rm -rf /var/lib/apt/lists/*

# Gemma4 (model_type="gemma4") is only available in the Transformers git dev
# branch. Installing from PyPI WILL fail at runtime with:
#   "The checkpoint you are trying to load has model type `gemma4` but
#    Transformers does not recognize this architecture."
# Keep this install line pointed at git+https until gemma4 lands in a release.
RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir \
        "torch>=2.4.0" \
        "git+https://github.com/huggingface/transformers.git" \
        "accelerate>=1.0.0" \
        "huggingface_hub" \
        "sentencepiece" "protobuf" \
        "gradio>=5.0" \
        "fastapi" "uvicorn" "httpx" "requests" \
        "Pillow" "PyMuPDF" "openai"

WORKDIR /app
COPY . /app

EXPOSE 7860

CMD ["python3", "app.py"]