Dockerfile

# Use NVIDIA CUDA base image with Python
FROM nvidia/cuda:12.1.0-cudnn8-runtime-ubuntu22.04

# Set working directory
WORKDIR /app

# Prevent interactive prompts during package installation
ENV DEBIAN_FRONTEND=noninteractive

# Install Python and essential build tools
RUN apt-get update && apt-get install -y \
    python3.10 \
    python3-pip \
    git \
    wget \
    curl \
    && rm -rf /var/lib/apt/lists/*

# Install uv - the fast Python package installer
RUN curl -LsSf https://astral.sh/uv/install.sh | sh
ENV PATH="/root/.cargo/bin:$PATH"

# Copy requirements for dependency installation
COPY requirements.txt .

# Install PyTorch with CUDA support using uv (much faster!)
RUN uv pip install --system torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121

# Install Diffusers from GitHub (latest version with Qwen support)
RUN uv pip install --system git+https://github.com/huggingface/diffusers

# Install other dependencies using uv
RUN uv pip install --system transformers accelerate safetensors pillow runpod

# Copy handler script
COPY handler.py .

# Optional: Pre-download the model (uncomment if you want to bake it into the image)
# This increases build time and image size but reduces cold start time
# RUN python3 -c "from diffusers import QwenImageEditPlusPipeline; QwenImageEditPlusPipeline.from_pretrained('Qwen/Qwen-Image-Edit-2511')"

# Set environment variables
ENV PYTHONUNBUFFERED=1

# Start the handler
CMD ["python3", "handler.py"]