FROM huggingface/transformers-pytorch-gpu:latest


RUN pip install --upgrade pip
RUN pip install transformers


ENV MODEL_NAME="Hadeel11/your-finetuned-model"
ENV USE_FP16=True

Expose the port for the inference API
EXPOSE 8080

Run the TGI server
CMD ["python3", "-m", "transformers_serve", "--model-id", "${MODEL_NAME}"]