Dockerfile

# Use standard Python image
FROM python:3.10

# Set working directory
WORKDIR /app

# Install wget
RUN apt-get update && apt-get install -y wget && rm -rf /var/lib/apt/lists/*

# Download the Model (Gemma 1B Q8)
RUN wget "https://huggingface.co/unsloth/gemma-3-1b-it-GGUF/resolve/main/gemma-3-1b-it-Q8_0.gguf" -O /gemma-3-1b-it-Q8_0.gguf

# Install FastAPI and Uvicorn
RUN pip install --no-cache-dir fastapi uvicorn pydantic sse-starlette

# CRITICAL STEP: Install pre-built llama-cpp-python for CPU
# This skips the "Building wheel" step that freezes your space
RUN pip install llama-cpp-python \
    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu

# Copy your main.py file
COPY main.py /app/main.py

# Give permissions
RUN chmod 777 /app/main.py

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