SHANGHAI — ACE ROBOTICS today announced that its open-source Kairos world model has achieved leading results across four global embodied-intelligence benchmarks: RoboTwin 2.0, LIBERO-Plus, WorldModelBench Robot and DreamGen. Kairos ranked first among evaluated world models and vision-language-action (VLA) systems on these benchmarks' public leaderboards as of 12 June 2026.
The project is openly available on GitHub, Hugging Face and ModelScope, providing researchers and developers with a public reference point for the model, benchmark results and technical materials.
Leading scene-level generalization on LIBERO-Plus
On LIBERO-Plus, a scene-level generalization benchmark evaluating robustness under seven real-world variables, Kairos achieved an overall score of 89.0, ranking first among all evaluated world models and VLA systems. It surpassed leading VLA models including ACoT-VLA (88.0), Pi 0.5 (85.7) and ProGAL-VLA (85.5). According to ACE ROBOTICS, this marks the first time a world-model approach has outperformed leading VLA systems on this benchmark.
Compact model with strong efficiency
On WorldModelBench Robot, Kairos-4B achieved an overall score of 9.30 with only 4 billion parameters, outperforming larger systems including the 28-billion-parameter Lingbot, 16-billion-parameter Cosmos 3, and 14-billion-parameter Abot-PhysWorld. It matched the top instruction-following score of Cosmos 3 with about one quarter of the parameters, a fourfold efficiency gain.
Unified architecture approach
ACE ROBOTICS attributes Kairos' performance to its native unified "multi-modal understanding-generation-prediction" architecture, which integrates these capabilities within a single backbone rather than stitching separate components together. The company describes Kairos-4B as the first embodied world model able to drive a physical robot directly on-device.
Commercial momentum
The results come as ACE ROBOTICS accelerates commercialization, having raised several hundred million U.S. dollars in the first half of 2026, including a recent Angel+ round backed by Geely Capital, Shenzhen Capital Group, and the Shanghai Sci-Tech Innovation Fund, with existing shareholder SenseTime's Guoxiang Capital increasing its stake.
