ByteDance Ltd. is developing an AI model specifically designed for real-time spatial video generation, positioning itself to compete with Meta Platforms Inc. and Alphabet Inc. in applications spanning robotics and autonomous systems. According to reports from Bloomberg, Founder Zhang Yiming is personally overseeing the development of this new model, which is set to unveil as early as October 2026. The billionaire has been coordinating efforts across business units and allocating AI resources and computing capacity to the project. However, as reported by Bloomberg, no specific timeline has been finalized and plans could change, highlighting the preliminary nature of these developments.
The model is built on Seedance — ByteDance's existing AI model for generating cinematic videos, allowing users to create interactive virtual worlds for live streams, short-form dramas and games. As reported by Bloomberg, the system can generate virtual worlds that respond to Pico headset users' voices or movements and offers on-demand videos with a latency of about 0.05 seconds at 20 frames per second. The model is rendered in the cloud at roughly 50 milliseconds of latency, which is the key strategic advantage - rendering spatial content remotely takes the computational load off the headset, which lowers what the hardware has to do and therefore what it has to cost. ByteDance owns Pico, its extended reality arm, and the model is reportedly meant to generate worlds that respond to Pico users' voices and movements.
According to Bloomberg, internal testing early in 2026 put ByteDance about 10% behind the global state of the art, on the same account as Google's Genie, which now lets users walk around Street View imagery rendered in real time. However, a launch next month would be ahead of that schedule, indicating accelerated development progress. The company is pursuing two routes at once: a vision-language-action approach aimed at embodied intelligence and robotics, and 3D simulation for entertainment and games. The spatial video model belongs to the second route, which is also the one with an existing user base attached to it, giving ByteDance a competitive advantage in the entertainment and gaming sectors. As reported by 36Kr, world models were said to command the company's largest data budget of any model direction, an eight-figure sum in renminbi that 36Kr's sources put at three to four times what rivals were spending.
The initiative is part of ByteDance's ongoing strategic shift toward generative AI, as the Beijing-based company redirects operations to compete with domestic rivals like Alibaba Group Holding Ltd. and Moonshot AI in frontier language models. According to Bloomberg, Zhang views the spatial-video model as a flywheel, linking ByteDance's AI models and cloud-computing resources with content platforms and Pico's hardware. The model aims to lower upfront VR adoption costs by moving computationally intensive spatial content generation to the cloud. As reported by Bloomberg, if it works, the contest over XR moves away from hardware specifications and towards models, cloud capacity and content distribution, three things ByteDance already has. For Meta and Apple, the implication is that a cloud-rendered world model does not beat a Vision Pro on fidelity, it just makes the fidelity somebody else's problem. Zhang Yiming hopes to use this to bring Douyin creators into the emerging field of "world models," competing with AI leaders such as Fei-Fei Li and Yann LeCun.
ByteDance has secured a $30 billion loan to expand AI capabilities and build data centers and hardware infrastructure, as reported by Bloomberg. The company has been weighing capital expenditure of as much as $70bn on its AI build-out, demonstrating significant commitment to AI development. ByteDance, founded by Zhang Yiming alongside Liang Rubo in 2012, is competing against Tencent Holdings Ltd. in the enterprise AI software market while catching up with domestic open-weight rivals in language model development. According to 36Kr, world models were said to command the company's largest data budget of any model direction, an eight-figure sum in renminbi that 36Kr's sources put at three to four times what rivals were spending. ByteDance is shifting a large portion of its B2B business toward video production and generation, having previously secured the substantial funding for AI capability expansion.
