GEN-0
$ 0
Trained on 270 000+ hours of real-world manipulation data; tested on semi-humanoid robots up to 16+ DoF
Showing 33–48 of 96 results


Trained on 270 000+ hours of real-world manipulation data; tested on semi-humanoid robots up to 16+ DoF


GEN-1 is a robotics foundation model designed to enable general-purpose intelligence and adaptable behavior across physical systems.


GEN-1.5 is an embodied foundation model that learns new manipulation tasks from 3-12 second demonstrations, generalizes across simulated and real robots, and recovers from errors through intelligent improvisation without requiring any code or fine-tuning.


GENE-26.5 is a general-purpose world model that predicts how environments change over time, enabling AI systems to reason, plan, and adapt across complex real-world scenarios.


Genie Envisioner 2.0 is AgiBot’s world model that acts as a learned simulator. It plays out robot tasks as video and judges whether each attempt succeeded.


GO-2 is an embodied AI foundation model that combines action reasoning with stable real-world execution.


GigaBrain-0.5M* is GigaAI’s robot action model that learns from a world model’s predictions. It reliably finishes long chores like folding laundry and making espresso.


GigaBrain-0.7 is GigaAI’s robot foundation model that lets robots preview outcomes with a built-in world model. It handles long, multi-step chores in homes and factories.


GO-3 Preview is AgiBot’s newest embodied foundation model. It helps robots understand tasks, predict the results of their actions and keep improving after deployment.


GR-Dexter is ByteDance Seed’s robot foundation model for dexterous hands. It controls two five-fingered hands to handle complex everyday tasks and copes well with new objects and instructions.


GR-RL is ByteDance Seed’s robot foundation model for precise, long manipulation tasks. It combines learning from demonstrations with real-world practice, and it is the first learning-based model to lace a shoe on its own.


GR00T 2 is NVIDIA’s open world action model for robots. It learns largely from human video, predicts how scenes will change and works across arms, dexterous hands and humanoids.


First VLA for full humanoid upper-body control at high frequency; dual-system architecture


Helix 02 is Figure AI’s in-house robot brain that controls the whole humanoid body at once, letting it walk, balance and handle objects as one smooth behavior.


HoloMotion-1 is Horizon Robotics’ open motion model that lets humanoid robots copy almost any human movement in real time, from dancing to kung fu.


Hy-Embodied-0.5-VLA is Tencent’s open robot action model. It turns images and instructions into precise two-handed movements and transfers to different robot arms.