ACE ROBOTICS' Kairos World Model Leads Multiple Global Embodied-Intelligence Benchmarks

SHANGHAI, CHINA - Media OutReach Newswire - 15 June 2026 - 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 the core capabilities of embodied intelligence, including complex robotic manipulation, scene-level generalization, physical-world modeling and zero-shot transfer. The project is openly available on GitHub, Hugging Face and ModelScope, giving researchers and developers a public reference point for the model, benchmark results and technical materials.

Tops Four Leading Benchmarks in World Generation & Prediction

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 by directly mapping perception and language inputs to robot actions, ACE ROBOTICS believes world models offer a more scalable path by explicitly learning the underlying dynamics of the physical world and predicting how environments evolve. Kairos is designed to validate that approach.

Leading scene-level generalization on LIBERO-Plus

One of Kairos' most significant results comes from LIBERO-Plus, a scene-level generalization benchmark proposed by the Shanghai Innovation Institute with Fudan University, Tongji University and the National University of Singapore. It evaluates robustness under seven real-world variables: camera angle, robot embodiment, language instruction, lighting, background, sensor noise and spatial layout.

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). It also showed strong environmental robustness, with near-ceiling performance on lighting (97.7), noise (96.8) and background (95.8), and ranked highly on camera angle and language instruction.

According to ACE ROBOTICS, this marks the first time a world-model approach has outperformed leading VLA systems on LIBERO-Plus for scene-level generalization, pointing to a path where robots adapt to homes, factories, retail spaces and other environments with far less environment-specific retraining.

A compact model with strong physical modeling efficiency

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, ranking first on the benchmark. With only 4 billion parameters, it outperformed larger systems including 28-billion-parameter Lingbot, 16-billion-parameter Cosmos 3, 14-billion-parameter Abot-PhysWorld and 5-billion-parameter Wan 2.2, setting a new record for parameter efficiency in embodied world models.

Kairos matched the top instruction-following score (2.36) of the 16-billion-parameter Cosmos 3 with about one quarter of the parameters, a fourfold efficiency gain. It scored 4.96 on physics adherence, with perfect marks on Newtonian mechanics and gravity, and a perfect score on temporal quality, reflecting strong temporal consistency and visual continuity over long horizons.

A unified architecture, not a modular pipeline

ACE ROBOTICS attributes Kairos' performance to its native unified "multi-modal understanding-generation-prediction" architecture. Unlike modular approaches that stitch together separate components for world understanding, generation and prediction, Kairos integrates these within a single backbone that shares one global world state, reducing the information loss and coordination latency between modules for more consistent physical modeling, stronger long-horizon prediction and more reliable action planning.

ACE ROBOTICS first introduced this architecture in December 2025, and the broader industry is now converging on a similar path: NVIDIA's Cosmos 3.0, introduced in 2026, adopts a comparable single-system design that brings vision reasoning, world generation and action prediction into one architecture. Built on this foundation, Kairos-4B is, in ACE ROBOTICS' description, the first embodied world model able to drive a physical robot directly on-device, closing the perception-to-action loop without intermediate translation latency.

Leading on synthetic data transfer and complex robot manipulation

Kairos also ranked first on DreamGen Bench, a benchmark led by NVIDIA with the University of Washington, UC Berkeley and UCLA that measures how well synthetic data generated by world models transfers to unseen objects, behaviors and environments, a key predictor of downstream robot-training value. Kair…