FROM pytorch/pytorch:2.6.0-cuda12.4-cudnn9-runtime

RUN pip install --no-cache-dir \
    "transformers>=5.3.0" \
    accelerate \
    soundfile \
    fastapi \
    "uvicorn[standard]" \
    huggingface_hub

COPY handler.py /app/handler.py
COPY serve.py /app/serve.py

WORKDIR /app
EXPOSE 5000
CMD ["uvicorn", "serve:app", "--host", "0.0.0.0", "--port", "5000"]