- RoboCasa, 24 tasks with the Panda Omron robot, average success rate, GR00T-N1.6-3B (examples/robocasa README) 66.2
- RoboCasa GR1 Tabletop Tasks, average of 24 tasks (examples/robocasa-gr1-tabletop-tasks README) 47.6
- SimplerEnv Bridge (WidowX) average 62.07%, GR00T-N1.6-bridge checkpoint. Fractal (Google robot) average 67.66% (examples/SimplerEnv README) 62.1
- LIBERO Spatial 97.65%, Goal 97.5%, Object 98.45%, 10 (Long) 94.35%, 200 trials per suite, no average printed (examples/LIBERO README)
- BEHAVIOR 50 tasks on the R1 Pro, average Task Progress 26.30% (π0.5 11.30%), a denser metric than the challenge Q score (examples/BEHAVIOR README)
GR00T N1.6
GR00T N1.6.
7
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
NVIDIA Dec 2025 GR00T N1.6
Open weights Flagship
No Epoch Capabilities Index score yet
- VLA-Arena 27.8% #8 /8
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| RoboCasa | 66.2% | #2 of 4 | Model card | |
| RoboCasa365 | 21.9% | #2 of 4 | Official leaderboard | |
| RoboCasa GR1 Tabletop Tasks | 47.6% | #3 of 4 | Model card | |
| RoboLab | 7.20% | #3 of 4 | Official leaderboard | |
| VLA-Arena | 27.8% | #8 of 8 | Official leaderboard | |
| SimplerEnv | 62.1% | #3 of 3 | Model card | |
| PhAIL | 72.3% | #2 of 2 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.