Unitree Robotics · 宇树科技
Embodied / robotics model — runs on robot hardware, not a token API. No per-1M pricing. For API-priced LLMs see the model catalog.
| Creator | Unitree Robotics (宇树科技) |
| Architecture | Vision-Language-Action (VLA) |
| Embodiment | Humanoid + manipulation |
| License | CC BY-NC-SA 4.0 |
UnifoLM-VLA is Unitree's open vision-language-action model, built to give humanoid robots physically grounded embodied intelligence. Published with a VLM base (UnifoLM-VLM-Base), a VLA base, and a LIBERO-benchmark variant. Released under CC BY-NC-SA 4.0 (non-commercial). Pairs naturally with Unitree's own humanoid hardware but the weights are open for research.
Humanoid robot control research; benchmarking VLA policies on LIBERO; teams already on Unitree hardware.
Your brand here — Reach our audience of professional directory owners and boost your sales.