We’ve just open-sourced all model checkpoints, training code, and evaluation code for RoboTwin, LIBERO-PRO, and our real-world experiments! 🚀
Everything is available here:
https://github.com/GehaoZhang6/Grounded-Action-Model
The Grounded Action Model team open-sources all model checkpoints, training code, and evaluation code for RoboTwin, LIBERO-PRO, and its real-world experiments on GitHub. The model builds robot foundation models on a pretrained 3D grounding model rather than language or video models.
We’ve just open-sourced all model checkpoints, training code, and evaluation code for RoboTwin, LIBERO-PRO, and our real-world experiments! 🚀
Everything is available here:
https://github.com/GehaoZhang6/Grounded-Action-Model
VLAs? WAMs? Are language and video models the right foundations for robotics? Introducing Grounded Action Model (GAM): a new paradigm that builds robot foundation models on top of a pretrained 3D grounding model. Ground first. Then learn to act. 🧵👇
Source: Jiafei Duan · x.comPublished · added here