VLA-Feedback Simulation Checkpoints

Official simulation checkpoints for VLA-Feedback, a vision-language-action model with feedback built on GR00T N1.5.

Folder	Trained for
`cp_sim_obs`	Dynamic tasks in RoboSuite
`cp_libero_obs`	LIBERO benchmark tasks (LIBERO-Object, LIBERO-Goal)

License

Apache 2.0. Derived from NVIDIA Isaac-GR00T.

Downloads last month: -; Downloads are not tracked for this model. How to track

Video Preview

Base model

Finetuned

(57)

this model