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, leading across core capabilities including complex robotic manipulation, scene-level generalization, physical-world modeling and zero-shot transfer.
Embodied intelligence faces a fundamental challenge: generalization. A robot must operate reliably in environments it has never seen, adapting to new lighting, layouts, objects, embodiments and noisy real-world conditions. While VLA models have become a prevailing approach, ACE ROBOTICS believes world models offer a more scalable path by explicitly learning the underlying dynamics of the physical world. Kairos is designed to validate that approach.
On LIBERO-Plus, a scene-level generalization benchmark proposed by the Shanghai Innovation Institute with Fudan University, Tongji University and the National University of Singapore, 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), as well as the Being-H0.7 world model (84.8). This marks the first time a world-model approach has outperformed leading VLA systems on this benchmark for scene-level generalization.
On WorldModelBench Robot, a physical-modeling benchmark proposed by researchers from UC Berkeley, UC San Diego, NVIDIA and MIT, Kairos-4B achieved an overall score of 9.30 with only 4 billion parameters, outperforming larger systems including 28-billion-parameter Lingbot, 16-billion-parameter Cosmos 3, 14-billion-parameter Abot-PhysWorld and 5-billion-parameter Wan 2.2. It matched the top instruction-following score (2.36) of Cosmos 3 with about one quarter of the parameters, a fourfold efficiency gain.
ACE ROBOTICS attributes Kairos' performance to its native unified architecture that integrates multi-modal understanding, generation and prediction within a single backbone. Unlike modular approaches, Kairos shares one global world state, reducing information loss and coordination latency. ACE ROBOTICS first introduced this architecture in December 2025, and NVIDIA's Cosmos 3.0, introduced in 2026, adopts a comparable single-system design.
Kairos also ranked first on DreamGen Bench, a benchmark led by NVIDIA with the University of Washington, UC Berkeley and UCLA, on both average physics adherence (AVG_PA 0.538) and overall average score (AVG_Score 0.618). On RoboTwin 2.0, a demanding dual-arm manipulation benchmark proposed by Shanghai Jiao Tong University and the University of Hong Kong with Shanghai AI Laboratory, Kairos scored 96.1%, a state-of-the-art result, ahead of VLA models such as G0.5 (93.2) and starVLA (88.3).
The results come as ACE ROBOTICS accelerates commercialization. The company has raised several hundred million U.S. dollars across financing rounds in the first half of 2026, including a recent Angel+ round backed by investors such as Dachen Caizhi, Shenzhen Capital Group and the Shanghai Sci-Tech Innovation Fund. The proceeds will support continued world-model research and integrated hardware-software solutions for sectors including smart retail, security and inspection, tourism and hospitality.
Kairos is openly available on GitHub, Hugging Face and ModelScope.
