Dockerfile

# Use a slightly more compatible base image
FROM python:3.10-slim

# Prevent Python from writing .pyc files
ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

# 🚀 Disable GPU build (CRITICAL FIX)
ENV CMAKE_ARGS="-DLLAMA_CUBLAS=off"

# Install only REQUIRED system dependencies
RUN apt-get update && apt-get install -y \
    gcc \
    g++ \
    make \
    cmake \
    libopenblas-dev \
    libgomp1 \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /code

# Upgrade pip first (important for wheel resolution)
RUN pip install --upgrade pip

# Copy requirements
COPY requirements.txt .

# 🚀 Force binary install (KEY FIX)
RUN pip install --no-cache-dir --prefer-binary -r requirements.txt

# Copy app
COPY . .

# Hugging Face uses port 7860
EXPOSE 7860

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