Dockerfile

FROM python:3.10-slim

# 1. Install Build Tools (REQUIRED for llama.cpp compilation)
# We added 'build-essential' and 'cmake' back to fix the build error.
RUN apt-get update && apt-get install -y \
    build-essential \
    cmake \
    curl \
    git \
    && rm -rf /var/lib/apt/lists/*

# 2. Install AI Engine
RUN pip install --no-cache-dir huggingface_hub "llama-cpp-python[server]>=0.2.82"

# 3. Setup User (Security)
RUN useradd -m -u 1000 user
USER user
ENV HOME=/home/user \
    PATH=/home/user/.local/bin:$PATH

WORKDIR $HOME/app

# 4. Download Model (Qwen 2.5 Coder 7B - Quantized)
RUN huggingface-cli download Qwen/Qwen2.5-Coder-7B-Instruct-GGUF \
    qwen2.5-coder-7b-instruct-q4_k_m.gguf \
    --local-dir . \
    --local-dir-use-symlinks False

# 5. Expose Port
EXPOSE 7860

# 6. RUN COMMAND (Lightweight Pro Config)
# Kept the optimized threading/context settings
CMD ["python3", "-m", "llama_cpp.server", \
     "--model", "qwen2.5-coder-7b-instruct-q4_k_m.gguf", \
     "--host", "0.0.0.0", \
     "--port", "7860", \
     "--n_threads", "2", \
     "--n_ctx", "16384", \
     "--n_batch", "512", \
     "--chat_format", "chatml"]