Dockerfile

FROM debian:bookworm-slim

RUN apt-get update && apt-get install -y --no-install-recommends \
    git build-essential cmake curl ca-certificates pkg-config \
    libcurl4-openssl-dev && rm -rf /var/lib/apt/lists/*

WORKDIR /app

RUN git clone --depth 1 https://github.com/ggerganov/llama.cpp.git \
 && mkdir -p build && cd build \
 && cmake -DCMAKE_BUILD_TYPE=Release \
          -DGGML_NATIVE=ON \
          -DLLAMA_BUILD_EXAMPLES=ON \
          -DLLAMA_BUILD_SERVER=ON \
          -DLLAMA_BUILD_TESTS=OFF \
          ../llama.cpp \
 && cmake --build . --target llama-server -j

RUN mkdir -p /models && \
    curl -fL -o /models/model.gguf \
    "https://huggingface.co/LiquidAI/LFM2-350M-ENJP-MT-GGUF/resolve/main/LFM2-350M-ENJP-MT-Q4_K_M.gguf?download=true"

EXPOSE 7860
CMD ["bash","-lc","/app/build/bin/llama-server -m /models/model.gguf --host 0.0.0.0 --port ${PORT:-7860} -c 2048 -ngl 0"]