Dockerfile

FROM python:3.12-slim

ENV PYTHONUNBUFFERED=1 PIP_NO_CACHE_DIR=1 PORT=7860 HF_HOME=/root/.cache/huggingface

WORKDIR /app

# faiss / numpy performance dep
RUN apt-get update && apt-get install -y --no-install-recommends libgomp1 && rm -rf /var/lib/apt/lists/*

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Optional: prefetch embedding model to reduce first-request latency
RUN python - <<'PY' || true
from sentence_transformers import SentenceTransformer
SentenceTransformer('Qwen/Qwen3-Embedding-0.6B')
PY

COPY . .

EXPOSE 7860

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