Dockerfile

# Use NVIDIA CUDA base image with Ubuntu 22.04, CUDA 12.2, and cuDNN 8
FROM nvidia/cuda:12.2.2-cudnn8-runtime-ubuntu22.04

# Set Python to run in unbuffered mode for better logging
ENV PYTHONUNBUFFERED=1

# Install Python 3.10, pip, and system dependencies in a single layer
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3.10 \
    python3-pip \
    libgl1 \
    && ln -s /usr/bin/python3.10 /usr/bin/python \
    && rm -rf /var/lib/apt/lists/*

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

# Copy application code
COPY main.py utils.py ./

# Verify the application can be imported without errors
RUN python -c "from main import app"

# Expose the application port
EXPOSE 7860

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