Moxinc Technology, founded by Zhejiang University PhD student Chen Tianrun, has unveiled MoWorld 2.0, a new general-purpose 4D world model. This iteration enhances the modeling of physical laws, integrating over ten physical elements like gravity, elasticity, and fluid dynamics to simulate object interactions and environmental changes more accurately. The model's output can be saved as 3D/4D Gaussian assets for further editing and use in applications such as game development, film production, and robotics simulation. Optimized for Huawei Ascend NPUs, MoWorld 2.0 achieves real-time inference speeds of up to 70 FPS with significantly reduced costs compared to GPU solutions, and has been integrated into end-user devices like smartphones. AI
IMPACT Enhances AI's ability to simulate complex physical interactions, potentially accelerating development in robotics, gaming, and autonomous systems.
RANK_REASON New model release from a prominent university lab with significant technological advancements and industry partnerships. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
- Chen Tianrun
- Huawei
- Huawei Ascend
- Kirin
- MoWorld 2.0
- Moxinc Technology
- World Internet Conference
- Zhejiang University
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
