Dockerfile

# Use a lightweight Python base
FROM python:3.10-slim

# Prevent interactive prompts
ENV DEBIAN_FRONTEND=noninteractive \
    PYTHONUNBUFFERED=1 \
    PYTHONDONTWRITEBYTECODE=1 \
    HF_HOME=/root/.cache/huggingface

# Set work directory
WORKDIR /code

# Install system dependencies (needed for PyTorch + FAISS + SentenceTransformers)
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    git \
    curl \
    libopenblas-dev \
    libomp-dev \
    && rm -rf /var/lib/apt/lists/*

# Install Python dependencies
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Pre-download Hugging Face models at build time
RUN python -c "from transformers import pipeline; pipeline('text-generation', model='NousResearch/Nous-Hermes-2-Mistral-7B-DPO')" \
 && python -c "from transformers import pipeline; pipeline('text2text-generation', model='google/flan-t5-large')" \
 && python -c "from sentence_transformers import SentenceTransformer; SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2')"

# Copy project files
COPY . .

# Expose port for FastAPI / Hugging Face Spaces
EXPOSE 7860

# Run FastAPI app with uvicorn
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]