AISphere Launches PixVerse R2, a Real-Time World Model That Remembers What Happens in a Session, Pandaily reports
Reported by Pandaily and Leiphone · not yet confirmed by the company or a second independent outlet. We update this page when it is.
AISphere launched PixVerse R2, a real-time world model that generates explorable audiovisual worlds from text, image, audio, and keyboard input while maintaining state across a session. It is now fully available on the web. The launch was reported by Pandaily and Leiphone.
China context
- Original name
- 爱诗科技
- Outside China
- Not stated in the sources yet
- Claims
- Company-reported; not yet independently evaluated
- For builders
- Developers outside China can monitor whether AISphere releases an API or open weights for PixVerse R2, which would enable integration into interactive applications.
- For investors
- Investors can track AISphere's user growth and any enterprise partnerships as indicators of commercial traction for real-time world models.
Translated from Chinese. Quotes and facts link to the original sources.
AISphere's PixVerse R2 generates explorable audiovisual worlds in real time, taking text, image, audio and keyboard input mid-stream and keeping state across a session. Users can enter the web version to experience action commands and interactive video-game-like interactions. The model was first proposed as a 'general real-time world model' direction in January 2026 with R1, and R2 is now fully launched.
PixVerse R2 separates capability and efficiency into two layers, allowing five types of growth—generation quality, long-range consistency, and multimodal control—to be scaled within a single framework. This addresses the tension between real-time speed and general capability that prevents simple scaling of video generation models.
AISphere's launch of PixVerse R2 puts pressure on other AI video and world model developers to match real-time interactivity and session memory, or risk losing users who expect persistent, controllable worlds rather than one-shot clips.
PixVerse R2 enables developers and creators to build interactive video experiences and games that respond to user input in real time, potentially reducing production costs and time compared to traditional game engines or pre-rendered video.
The next verifiable signal is whether AISphere releases technical details or benchmarks for PixVerse R2's long-range consistency and multimodal control, and whether third-party developers build interactive experiences on top of it.