Movens-Zero: Scaling Humanoid Motion Foundation Models with Large-Scale Human Videos
Technical report
Authors & affiliations (24)
Author order: Original order
- First published
- Not provided
- Latest paper revision
- v1 · 2026-10-30
- Venue
- arXiv
Action expert: 187.9M, 4 blocks, width 1152, 18 attention heads. Movens-Zero is a humanoid motion foundation model. It uses Qwen2.5-Omni-3B and a flow-matching action expert to turn text, speech, music and video into Unitree G1 whole-body motion references. The project provides training, inference, simulation and robot deployment code.
View full model descriptionTechnical report
Author order: Original order
SourceVerified on · 2026-11-19
Some training datasets are public; this does not establish access to the complete training mixture.
| File | Access |
|---|---|
| Movens-Zero M · step 35000 | Access |
SourceVerified on · 2026-11-19
Available weights or code do not establish plug-and-play robot compatibility. Check the required configuration, data and interfaces in the source documentation.
action_generation
Unitree G1 whole-body reference trajectories
5000000 sequences
https://huggingface.co/datasets/wendell0218/Movens-5M
Companion dataset with five million physically executable humanoid motion sequences.
5000000 sequences · other
Supported · Reported
Unitree G1 EDU, 29 actuated joints. Official deployment uses a GPU inference server and the GR00T whole-body controller; the real-robot server enforces a 180 ms per-chunk deadline.
Action expert: 187.9M, 4 blocks, width 1152, 18 attention heads.
Weight upload: 2026-10-25 19:03:51 UTC+8. Training step: 35,000.
The safetensors header contains 4,891,355,152 stored tensor elements across the backbone and action expert.
The project has not specified a code or checkpoint license. The Qwen2.5-Omni-3B backbone has its own Qwen Research License.
Paper affiliations: Zhejiang University, Unitree Robotics, Joint Laboratory of Embodied Intelligence, ZJU & Unitree Robotics, Bit Oasis, HiThink Research, Microsoft Research, East China Normal University, Peking University, National University of Singapore.
Cover: Figure 7 by Bu et al., arXiv:2610.41877, CC BY 4.0. Converted to WebP.
7ac8742c9bcd · 18943283944 bytes · SHA-256 a01d56d42e4fec4d471b0696001e4b4ccea39698c2b3afc2180b4eb4dfa15a29
Deployment evidence points to the 29-DOF Unitree G1 EDU configuration. The authors provide a real-robot deployment path, with no separate on-robot validation disclosed for every checkpoint.
