Dockerfile

FROM ghcr.io/ggml-org/llama.cpp:full-cuda

WORKDIR /app

RUN apt update && apt install -y python3 python3-pip python3-venv
RUN python3 -m venv /opt/venv
ENV PATH="/opt/venv/bin:$PATH"

RUN pip install -U pip huggingface_hub

RUN python3 -c 'from huggingface_hub import hf_hub_download; \
    repo="HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive"; \
    hf_hub_download(repo_id=repo, filename="Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-Q5_K_P.gguf", local_dir="/app"); \
    hf_hub_download(repo_id=repo, filename="mmproj-Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-f16.gguf", local_dir="/app")'

CMD ["--server", \
     "-m", "/app/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-Q5_K_P.gguf", \
     "--mmproj", "/app/mmproj-Gemma-4-E4B-Uncensored-HauhauCS-Aggressive-f16.gguf", \
     "--host", "0.0.0.0", \
     "--port", "7860", \
     "-t", "2", \
     "-c", "128000", \
     "-n", "38912", \
     "--n-gpu-layers", "99"]