Dockerfile

# 🏗️ Stage 1: Build dependencies in a lightweight Python image
FROM python:3.9-slim AS builder

WORKDIR /app

# Install dependencies in a single step to optimize layers
COPY requirements.txt . 
RUN pip install --no-cache-dir -r requirements.txt \
    && pip install --no-cache-dir torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu \
    && pip install --no-cache-dir uvicorn[standard] transformers

# Verify installation
RUN python -c "import torch; print(torch.__version__); print(torch.cuda.is_available())"

# 🚀 Stage 2: Final optimized production image
FROM python:3.9-slim

WORKDIR /app

# Set up a persistent cache for Hugging Face models
RUN mkdir -p /app/cache && chmod -R 777 /app/cache
ENV HF_HOME=/app/cache

# Copy dependencies from builder stage
COPY --from=builder /usr/local/lib/python3.9/site-packages /usr/local/lib/python3.9/site-packages
COPY --from=builder /usr/local/bin /usr/local/bin

# Create non-root user for security
RUN useradd --create-home appuser
USER appuser

# Copy application code
COPY . .

# Expose port
EXPOSE 7860

# Start FastAPI with optimized settings
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860", "--workers", "1", "--timeout-keep-alive", "10"]