UniMate
by Friedrich-M
UniMate is a research model that generates 3D character animation from a text description, and it works across many different skeleton types such as people, animals, birds and insects. It comes with training code, preview checkpoints, a large motion dataset and a tool for preparing your own rigged models.
- Stars
- 2.2k
- Licence
- MIT
- Runs on
- A gpu machine, with a GPU
- Setup
- Technical
- Language
- Python
- Age
- 96 days
What you can do with it
- Generate animation clips for a rigged character from a text prompt
- Prepare your own GLB, glTF or FBX rigged models so the model can animate them
- Train or fine-tune the model on the UniML3D motion dataset
Make money with it
A freelancer could offer quick animated versions of clients' rigged characters or creatures for games and videos, using the generated motion as a starting point.
An idea, not financial advice. Check the licence before charging for anything built on it.
Ways a business could use it
- A small game studio could draft animations for non-human characters quickly, cutting the cost of buying or hand-making motion clips.
- A freelance 3D artist could produce first-pass animations for client characters before refining them by hand.
Can you use it commercially?
Free to use, change and sell, including in paid products. Keep the copyright notice.
Not legal advice. Read the project's licence file before building a product on it.
Getting it running
Create a conda environment with Python 3.10, install the packages with pip from requirements.txt, then download checkpoints and data from Hugging Face.
Follow the official instructions in the project README.
Things to check
- Academic research code with preview checkpoints, so results may be inconsistent
- Needs a GPU and a technical setup
- Part of the training data (Truebones ZOO) is a paid pack that must be bought separately
- Training data includes Mixamo and Objaverse assets, so check their licences before commercial use
Who's talking about it
Checked 27 Oct 2026. Summary drafted by AI from the project's README.
