Dockerfile

# Use PyTorch with CUDA 12.1 for L4 GPU compatibility
FROM pytorch/pytorch:2.2.2-cuda12.1-cudnn8-runtime

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        git \
        git-lfs \
        build-essential \
        ffmpeg \
        curl \
        wget && \
    rm -rf /var/lib/apt/lists/*

# Create non-root user for Hugging Face Spaces
RUN useradd -m -u 1000 user
USER user
WORKDIR /app

# Set environment variables
ENV PATH="/home/user/.local/bin:$PATH"
ENV HF_HOME=/app/.cache
ENV PYTHONPATH="/app:$PYTHONPATH"

# Copy application files
COPY --chown=user . /app

# Upgrade pip and install Python dependencies
RUN pip install --upgrade pip

# Install requirements
COPY --chown=user requirements.txt .
RUN pip install -r requirements.txt # --no-cache-dir 

# Install additional optimizations for L4 GPU
RUN pip install --no-cache-dir flash-attn --no-build-isolation || echo "Flash attention not available, continuing..."

# Expose port for Hugging Face Spaces
EXPOSE 7860

# Environment variables for the application
ENV PYTHONUNBUFFERED=1
ENV UVICORN_HOST=0.0.0.0
ENV UVICORN_PORT=7860

# Health check using FastAPI health endpoint
HEALTHCHECK --interval=30s --timeout=30s --start-period=60s --retries=3 \
    CMD curl -f http://localhost:7860/health || exit 1

# Start the FastAPI application with uvicorn
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]