---
title: DMAD Video Studio
emoji: 🎞️
colorFrom: indigo
colorTo: purple
sdk: static
app_file: index.html
short_description: Explore and try DMAD video generation
models:
- ZhengmingYu/DMAD
---

# DMAD Video Studio

A lightweight, unofficial browser interface to explore the original DMAD MiniMax-H3 examples and try text-to-video through Hugging Face's WaveSpeed Inference Provider. This static Space does **not** host model weights or run an inference server. It routes to WaveSpeed's MiniMax-H3 text-to-video LoRA endpoint and requests the project's 4-step full-critic adapter. Exact compatibility, provider sampler settings, pricing and output are unverified until tested against the real service.

The sample videos and workflows are linked from [ZhengmingYu/DMAD](https://huggingface.co/ZhengmingYu/DMAD); credit to the DMAD authors and the MiniMax-H3 project. Check the [MiniMax H3 Community License](https://huggingface.co/ZhengmingYu/DMAD/blob/main/LICENSE) and provider terms before use. WaveSpeed may bill the account associated with your inference token; pricing, availability and billing terms may change. A token with Inference Providers permissions is needed. The browser sends the token to Hugging Face's router and the prompt to the provider; the app does not intentionally store the token.