From the source
Odyssey released Odyssey-3, a learned dynamical system implemented as an autoregressive diffusion transformer that predicts object motion and interaction.
The Pro variant achieved a state-of-the-art score of 66.1 on the Physics-IQ Verified video-to-video benchmark.
The model generates embodied environments in real time from prompts and can be adapted to control robot arms, humanoids, and vehicles with limited training data.
