Dockerfile

# Use an official Python runtime as the base image
FROM python:3.10-slim

# Set environment variables for non-interactive installation
ENV DEBIAN_FRONTEND=noninteractive

# Install system dependencies and other tools
RUN apt-get update && apt-get install -y \
    git \
    cmake \
    build-essential \
    libomp-dev \
    python3-dev \
    wget \
    curl \
    libtorch-dev \
    libboost-all-dev \
    git-lfs \
    && rm -rf /var/lib/apt/lists/*

# Install Git LFS and initialize it
RUN git lfs install

# Set the working directory
WORKDIR /content

# Step 1: Clone and set up ExecuTorch
RUN git clone https://github.com/pytorch/executorch && \
    cd executorch && \
    git checkout dfbf6fd53546eb86e18f2e5cc693d70a82e3b03f && \
    git submodule sync && \
    git submodule update --init

# Step 2: Install ExecuTorch requirements (verify and handle errors)
RUN python3 -m pip install --upgrade pip && \
    pip install torch && \
    if [ -f "/content/executorch/requirements.txt" ]; then \
        python3 -m pip install --no-cache-dir -r /content/executorch/requirements.txt; \
    else \
        echo "No requirements.txt found, skipping..."; \
    fi

# Step 3: Install ExecuTorch dependencies with error handling
RUN cd /content/executorch && \
    if [ -f "install_requirements.sh" ]; then \
        bash ./install_requirements.sh --pybind || echo "Warning: Failed to execute install_requirements.sh"; \
    else \
        echo "install_requirements.sh not found, skipping..."; \
    fi && \
    cd /content/executorch/examples/models/llama && \
    if [ -f "install_requirements.sh" ]; then \
        bash ./install_requirements.sh || echo "Warning: Failed to execute llama/install_requirements.sh"; \
    else \
        echo "llama/install_requirements.sh not found, skipping..."; \
    fi

# Step 4: Clone the model repository from Hugging Face
RUN git clone https://huggingface.co/executorch-community/Llama-3.2-1B-Instruct-SpinQuant_INT4_EO8-ET /content/llama-model && \
    cd /content/llama-model && \
    git lfs pull

# Step 5: Install FastAPI and other necessary Python packages
RUN python3 -m pip install fastapi uvicorn pydantic

# Step 6: Copy the FastAPI application and model runner script into the container
COPY app.py /content/app.py

# Expose the port FastAPI will run on
EXPOSE 7860

# Step 7: Set up the entry point to run the FastAPI server
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]