LingBot-VLA 2.0 Brings Causal World Modeling To Open-Source Robotics
Posts on X are touting an open robotics model that backers say can run on standard hardware and handle whole-body control across many robot platforms.
621575.6K
“What I care about most: this ships as a real foundation, open and reproducible, with a paper, code, and weights you can run today. That is how the whole field moves faster.”
Robert Scoble@ScobleizerTECHStory Brief
Researchers and robotics watchers on X are circulating LeRobotHF's announcement of LingBot-VLA 2.0, an open-source vision-language-action model described in those posts as bringing causal world modeling into robotics. Supporters say it unifies visual dynamics prediction with action inference, supports whole-body control across 20 robot hardware platforms, and uses a sparse mixture-of-experts design that helps it run efficiently enough for practical deployment. For now, though, the public evidence in view is still the announcement posts and follow-on commentary on X.
“On a standard 4090??? Now that is cool. For robotics to truly work, the models need to run at the edge.”
KuphDevTECH