Ant Lingbo

“Released 6 open-source models in their second generation: Limbo Viren (geometric vision pre-training), Limbo Depth (spatial perception), VLA 2.0 (cross-embodiment action model), Video (MoE-native video generation), Word (causal pre-training), and VA 2.0 (the integrated robot-native foundation model).”

“In 26 this year, or rather this half-year, we made what I think is a fairly big decision: since we can't push the digital-world teams to change, we'll redo everything from scratch based on the physical world's needs.”

“We built our pre-training purely from a geometric angle — you can roughly understand it as points, lines, and surfaces. Starting from edges, we let the model better understand the concept of geometry.”

“The robot data is about two orders of magnitude smaller than internet data. Even with our 60,000 hours this time — honestly compared to last time, it's not an order-of-magnitude improvement.”