ByteDance preps real-time spatial video world model, Zhang Yiming personally overseeing, Bloomberg reports
智东西 · wechat · 2026-09-08
ByteDance is preparing to launch an AI model for real-time spatial video generation as soon as next month, with founder Zhang Yiming personally coordinating AI resources and compute across business units, Bloomberg reports citing sources.
- Built on ByteDance's Seedance video model, it aims to create interactive virtual worlds for livestreaming, short dramas, and games, pushing ByteDance into the emerging "world model" field alongside Google's Genie
- Positioned as a flywheel linking ByteDance's AI models, cloud, content platforms, and PICO XR hardware: generating video responsive to headset users' voice and motion at 0.05s latency and 20fps
- Offloading generation to the cloud would cut PICO headset compute needs and cost, paving the way for lighter hardware; PICO Space Pro was delayed to Q4 and appears to have passed FCC certification
- Bloomberg sees VR competition shifting from hardware specs to AI models, compute, and content distribution; ByteDance recently secured $29.6B in loans for AI expansion
Related event: ByteDance Secretly Builds Real-Time World Model Under Zhang Yiming(4 posts)→
More from Embodied
- NVIDIA releases MotionBricks: real-time humanoid motion generation at 15,000 FPS with 2ms latency — _akhaliq · 2026-10-26
- Claude Code finds root backdoor in Honda Civic head unit, then debloats it — IndraVahan · 2026-10-26
- Temple health wearable opens pre-orders, users ask: why wear it before benefits are proven — abhishekcode42 · 2026-10-26
- FLASH: GPU-native deformable sim learns cloth folding in minutes on one RTX 5090 — ChongZzZhang · 2026-10-26
- Luke Wroblewski spots it: iPhone Duo's corner radius mirrors the original iPhone — LukeW · 2026-10-26
- Humanoids shattered forecasts: half-marathon record in 2026, 100m faster than Bolt — ben_j_todd · 2026-10-26