Genie Operator-2
$ 0
GO-2 is an embodied AI foundation model that combines action reasoning with stable real-world execution.
Showing 17–32 of 40 results
-
-
Helix
$ 0
First VLA for full humanoid upper-body control at high frequency; dual-system architecture
-
Interactive World Simulator (IWS)
$ 0
Action-conditioned world model generating 10+ min of interactive predictions at ~15 FPS on RTX 4090; supports teleoperation and policy evaluation
-
IRIS
$ 0
Imagination-based agent using a transformer world model for RL; strong sample efficiency in simulated environments
-
-
MolmoAct
$ 0
Multimodal VLA from the Molmo family adapted for robot action prediction; open weights with strong zero-shot generalisation
-
NeRD
$ 0
Neural Radiance Dynamics: combines NeRF with physics simulation for photorealistic robot manipulation world modelling
-
NVIDIA Cosmos
$ 0
Physical AI world foundation model for generating photorealistic video simulations; supports robot and AV training
-
-
-
-
OpenVLA
$ 0
Open-source 7B VLA trained on Open X-Embodiment dataset; widely adopted open baseline for robot manipulation
-
Psi R1
$ 0
Psi R1 is a Vision-Language-Action (VLA) model by PsiBot that enables robots to perform complex, long-horizon tasks through integrated perception, reasoning, and action.
-
Rhoda AI FutureVision
$ 0
Proprietary predictive world model for autonomous robot decision-making; anticipates consequences of robot actions
-
RLDX-1
$ 0
RLDX-1 is a robotics foundation model from RLWRLD that specializes in dexterous robotic manipulation and humanoid interaction. The model combines visual understanding, language processing, tactile sensing, and force feedback to help robots perform complex real-world tasks with high precision.
-
Sanctuary AI Carbon
$ 0
Proprietary full-stack AI for Phoenix humanoid; not a separable foundation model
