VLA models
NVIDIA Isaac GR00T N1.7
United States
NVIDIA's flagship open robot foundation model line, started with GR00T N1 (first open humanoid foundation model, March 2025). N1.7 (GA April 2026) swaps in a Cosmos-Reason2-2B backbone and a relative end-effector action space shared between robot and human embodiments.
OrganizationNVIDIA
CountryUS
Release2026-04
Params (B)3
Open weightsyes
Architecturedual-system VLA: Cosmos-Reason2-2B VLM (System 2) + 16-layer diffusion-transformer action head (System 1), 40-step action horizon, relative end-effector action space
LicenseNVIDIA Open Model License (weights); Apache-2.0 (code)
Modalitiesvision, language, action
Embodimentshumanoid, manipulator, cross-embodiment
Training dataSeveral thousand hours of teleoperated demos (bimanual YAM, AgiBot Genie-1, simulated Galaxea R1 Pro, Unitree G1 locomanipulation) plus 20,000 h of EgoScale egocentric human video
Sources : NVIDIA (GitHub) (2026-07-09)NVIDIA Research (GEAR Lab) (2025-12-15)arXiv (2025-03-18)
← Back to comparator · VLA models