---
title: Sulphur 2 Demo
emoji: 🌋
colorFrom: yellow
colorTo: red
sdk: gradio
sdk_version: 6.26.0
app_file: app.py
python_version: "3.12"
startup_duration_timeout: 1h
short_description: Text and image to video with Sulphur 2
models:
  - SulphurAI/Sulphur-2-base
  - diffusers/LTX-2.3-Distilled-Diffusers
tags:
  - text-to-video
  - image-to-video
  - zero-gpu
  - ltx-2.3
---

# Sulphur 2 Demo

Interactive **text-to-video** and **image-to-video** demo for
[SulphurAI/Sulphur-2-base](https://huggingface.co/SulphurAI/Sulphur-2-base),
an LTX 2.3 fine-tune with synchronized audio.

The Space loads the [Diffusers conversion](https://huggingface.co/CalamitousFelicitousness/LTX-2.3-Sulphur2-Distilled-Diffusers)
of the Sulphur 2 distilled transformer, plus the ungated
[LTX-2.3 Distilled Diffusers](https://huggingface.co/diffusers/LTX-2.3-Distilled-Diffusers)
text encoder, VAE, and vocoder. Generation uses the official distilled 8-step schedule.

## Tips

- Write one cinematic shot: subject, motion, camera, lighting, and sound.
- Optional first-frame image enables image-to-video.
- Keep clips short; GPU time comes from the visitor's ZeroGPU quota.

Weights follow the [LTX-2 Community License](https://huggingface.co/SulphurAI/Sulphur-2-base/blob/main/LICENSE.txt).