Dockerfile

FROM python:3.10-slim
# BUILD_TS=1780158810

# Configure environments
ENV DEBIAN_FRONTEND=noninteractive
ENV PYTHONUNBUFFERED=1
ENV PORT=7860

# Install system dependencies
RUN apt-get update && apt-get install -y \
    wget \
    curl \
    unzip \
    && rm -rf /var/lib/apt/lists/*

# Set up workdir
WORKDIR /workspace

# Download and install llama.cpp pre-compiled Ubuntu binary release
RUN mkdir -p bin models bin_temp && \
    curl -L -o llama.tar.gz https://github.com/ggml-org/llama.cpp/releases/download/b9279/llama-b9279-bin-ubuntu-x64.tar.gz && \
    tar -xzf llama.tar.gz -C bin_temp --strip-components=1 && \
    cp -r bin_temp/* bin/ && \
    chmod +x bin/llama-server && \
    rm -rf llama.tar.gz bin_temp

# Copy requirements and install
COPY requirements.txt requirements.txt
RUN pip install --no-cache-dir -r requirements.txt

# Copy all codebase
COPY . .

# Run download model script in build phase to embed the model in the image
RUN python download_model.py

# Create a non-root user for security (required by Hugging Face)
RUN useradd -m -u 1000 user
RUN chown -R user:user /workspace
USER user

# Set environment variables for the app
ENV LLAMA_SERVER_PATH=/workspace/bin/llama-server
ENV LD_LIBRARY_PATH=/workspace/bin:$LD_LIBRARY_PATH
ENV LLAMA_PORT=8080
ENV FASTAPI_PORT=7860
ENV LLAMA_HOST=127.0.0.1
ENV LLAMA_THREADS=2

# Expose port
EXPOSE 7860

# 
# ── Agentic LLM System ──
# Install optional agent tool dependencies
RUN pip install --no-cache-dir aiohttp>=3.9.0 2>/dev/null || true


CMD ["python", "main.py"]