MoCore unveils MoWorld 2.0, a 4D world model natively built on Huawei Ascend for embodied AI training and simulation
Overview
MoCore launched MoWorld 2.0 at the World Internet Conference, featuring physics-based interaction and real-time rendering up to 70 FPS on domestic Ascend hardware. The system supports robotics training in virtual environments with reduced inference costs compared to GPU solutions.
Generated from attributed reports · Updated 2 hours ago
Event evidence and corrections
1 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · video inference frames per second: 70 other · Basis not reported
“官方表示，MoWorld 2.0 在实时性能上同步升级： 视频推理： 由 1.0 版本最高约 50 帧每秒，提升至约 70 帧每秒； 空间资产渲染： 3D/4D 高斯资产渲染速度可达 500 帧每秒以上； 手机端渲染： 相关测试可达到约 100 帧每秒。”
Exact source · revision 2ithome
Reported quantity · spatial asset rendering frames per second: 500 other · Basis not reported
“官方表示，MoWorld 2.0 在实时性能上同步升级： 视频推理： 由 1.0 版本最高约 50 帧每秒，提升至约 70 帧每秒； 空间资产渲染： 3D/4D 高斯资产渲染速度可达 500 帧每秒以上； 手机端渲染： 相关测试可达到约 100 帧每秒。”
Exact source · revision 2ithome
Reported quantity · mobile device rendering frames per second: 100 other · Basis not reported
“官方表示，MoWorld 2.0 在实时性能上同步升级： 视频推理： 由 1.0 版本最高约 50 帧每秒，提升至约 70 帧每秒； 空间资产渲染： 3D/4D 高斯资产渲染速度可达 500 帧每秒以上； 手机端渲染： 相关测试可达到约 100 帧每秒。”
Exact source · revision 2ithome
Reported quantity · video inference frames per second (previous): 50 other · Basis not reported
“官方表示，MoWorld 2.0 在实时性能上同步升级： 视频推理： 由 1.0 版本最高约 50 帧每秒，提升至约 70 帧每秒； 空间资产渲染： 3D/4D 高斯资产渲染速度可达 500 帧每秒以上； 手机端渲染： 相关测试可达到约 100 帧每秒。”
Exact source · revision 2ithome
Report timeline
Follow attributed reports and material updates.
- IT Home — RoboticsMoCore unveils MoWorld 2.0, a 4D world model natively built on Huawei Ascend for embodied AI training and simulation
MoCore launched MoWorld 2.0 at the World Internet Conference, featuring physics-based interaction and real-time rendering up to 70 FPS on domestic Ascend hardware. The system supports robotics training in virtual environments with reduced inference costs compared to GPU solutions.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).