Sulphur-2 Inference Endpoint

Custom-handler repository for deploying SulphurAI/Sulphur-2-base (an LTX-2.3 fine-tune) on Hugging Face Inference Endpoints.

The actual weights are not stored here. handler.py pulls only the files it needs from:

  • SulphurAI/Sulphur-2-base β€” distilled checkpoint + distill LoRA
  • Lightricks/LTX-2.3 β€” spatial upscaler
  • google/gemma-3-12b-it-qat-q4_0-unquantized β€” text encoder (gated, requires HF_TOKEN)

See NOTES.md for deployment instructions, hardware requirements, and the call protocol.

Files

  • handler.py β€” EndpointHandler for the Default container
  • requirements.txt β€” best-effort pip install path (see NOTES.md for caveats)
  • Dockerfile β€” Custom Container build (recommended path)
  • entrypoint.sh β€” container entrypoint for the Custom Container
  • NOTES.md β€” deployment + usage
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support