Motus
modelTSAIL's unified latent-action world model spanning video prediction and robotics (1.2K★ in its first month) — the lab's bid to merge its diffusion-video and robot-FM lines into a single action-conditioned world model.
TSAIL's unified latent-action world model spanning video prediction and robotics (1.2K★ in its first month) — the lab's bid to merge its diffusion-video and robot-FM lines into a single action-conditioned world model.