Add handler.py for Hugging Face Inference Endpoints deployment
#3
by SaiPavankumar22 - opened
Hi NVIDIA Team,
Thank you for releasing NVIDIA-NemotronLabs-VoiceChat-11B!
Currently, the repository does not include a custom handler.py file required for custom pipeline execution on Hugging Face Inference Endpoints. Because of this, deploying the model directly via HF Inference Endpoints fails or requires a custom container setup.
Could you please consider adding a standard EndpointHandler implementation (handler.py) or providing documentation/examples on how to serve this model using HF Inference Endpoints?
Thank you!