# EVO1

EVO1 is a Vision-Language-Action policy for robot control. The LeRobot
integration uses an InternVL3 vision-language backbone with a flow-matching
action head, and supports staged training through the standard LeRobot policy
APIs.

The upstream EVO1 project is available at
[MINT-SJTU/Evo-1](https://github.com/MINT-SJTU/Evo-1).

```bibtex
@misc{evo1,
  title = {EVO1},
  author = {{MINT-SJTU}},
  year = {2026},
  howpublished = {\url{https://github.com/MINT-SJTU/Evo-1}},
}
```