DMAD (4-step MiniMax H3 students)
by Zhengming Yu et al.
Four-step LoRA students of MiniMax H3 for fast joint audio-video generation.
AI video generators and platforms Open source
1. Price: huggingface.co/ZhengmingYu/DMAD
Open weights, free to download; self-hosted. Needs the MiniMax H3 base model.
huggingface.co/ZhengmingYu/DMADNo quote yet: price is read from the official pages.
2. Commercial use: huggingface.co
“You may not use, reproduce, modify, distribute, or display the MiniMax H3 Works or any of their Outputs or results outside the Applicable Territory.”
Show "MiniMax H3" in the interface of commercial products; ask MiniMax for a license above 20 million USD yearly revenue.
huggingface.co3. Credit the author: huggingface.co
“on the user interface of commercial product or service that uses MiniMax H3 or MiniMax H3 Works.”huggingface.co
4. Agent access: github.com/Yzmblog/DMAD
There is no package or API. Files are downloaded by hand from the official site.
github.com/Yzmblog/DMADLicense checked Oct 27, 2026 against 1 official page. Tap a verdict to see its source. How we checked Report an error
DMAD (4-step MiniMax H3 students) popularity
License
MiniMax H3 Community License Agreement.
“You may not use, reproduce, modify, distribute, or display the MiniMax H3 Works or any of their Outputs or results outside the Applicable Territory.” huggingface.co
“on the user interface of commercial product or service that uses MiniMax H3 or MiniMax H3 Works.” huggingface.co
“you must, before making that product or service available and throughout its operation, implement, maintain, test, and periodically review reasonable and proportionate technical and organizational safeguards” huggingface.co
“MiniMax claims no rights over the Outputs you generate.” huggingface.co
“You may not use the MiniMax H3 Works or any of their Outputs or results to improve any other artificial intelligence model” huggingface.co
- Use is allowed only inside the Applicable Territory: it excludes the EU, UK, South Korea and the USA.
- Show "MiniMax H3" in the interface of commercial products; ask MiniMax for a license above 20 million USD yearly revenue.
- Do not use the weights or their outputs to train or improve another AI model.
- Hosting generation for third parties needs safeguards under the Acceptable Use Policy.
AllowedWith conditionsNot allowedNot stated or not checked yet
Is DMAD (4-step MiniMax H3 students) free for commercial use? The full answer Not legal advice. Check the license page before you ship.
For your agent
License rules tell your agent what it may do with DMAD (4-step MiniMax H3 students). Paste them into your prompt or AGENTS.md.
- DMAD (4-step MiniMax H3 students) (AI video generators and platforms). Four-step LoRA students of MiniMax H3 for fast joint audio-video generation. Price: free; commercial use with conditions; credit required; ship inside your app with conditions; no AI training on its content.
Connect: download from https://huggingface.co/ZhengmingYu/DMAD (no API).
Rule: Use is allowed only inside the Applicable Territory: it excludes the EU, UK, South Korea and the USA.
Rule: Show "MiniMax H3" in the interface of commercial products; ask MiniMax for a license above 20 million USD yearly revenue.
Rule: Do not use the weights or their outputs to train or improve another AI model.
Rule: Hosting generation for third parties needs safeguards under the Acceptable Use Policy.
License: https://huggingface.co/ZhengmingYu/DMAD/blob/main/LICENSE (checked 2026-10-27; details: https://agentgid.com/assets/dmad/)
Alternatives
DMAD (4-step MiniMax H3 students) This page
Stable Video Diffusion (SVD-XT)
Image-to-video model that turns one still frame into a short 25-frame clip at 576x1024.
Open-source video toolkit for AI agents that clones reference videos into variants, plus a hosted multi-model video API.
Open MoE video model family for text-to-video, image-to-video, speech-to-video and character animation at 480p/720p.
Open-source platform that automates AI short-drama production from script and characters to storyboard and video compositing.
Open-source text-to-video and image-to-video generation models (CogVideo, CogVideoX) with inference and fine-tuning code.
Agents that do the same job
From AgentGid's agent catalog, not yet checked as asset sources: their license and price are on their agent pages.