Zhang Yiming personally oversees ByteDance's world model: Seedance base generates interactive worlds in real time.

Beating AI News (from Dongcha) – Bloomberg, citing people familiar with the matter, reports that ByteDance is developing a real-time spatial video model, set to launch as early as October, with founder Zhang Yiming personally overseeing the project. Built on Seedance, the model aims to respond to users’ voices and actions in real time, creating an interactive virtual world on the fly. The model is planned to be integrated with Pico headsets. Sources add that it can generate content in the cloud at roughly 20 frames per second, with a latency of about 0.05 seconds. Offloading most computing to the cloud reduces the computational power required for the headsets themselves. Beyond VR, ByteDance also plans to apply the model to live streaming, short dramas, and games. This is not a sudden pivot to world models for ByteDance: 36Kr reported back in June that world models are one of ByteDance’s four key AI priorities for 2026, with the goal of launching at least one version by the end of the year to compete with Google’s Genie 3. Genie 3 can already generate 720p interactive worlds at 20–24 frames per second and run continuously for several minutes.