SimplerEnv Google Robot (Visual Matching) leaderboard

Real-to-sim replica of the RT-1 Google Robot setup, with the simulated scene overlaid on real backgrounds to match the real camera view. Policies are usually fine-tuned on the RT-1 (Fractal) data. Scores are the mean of three tasks; many papers add a fourth (place apple in drawer) to their average, so RoboFYI recomputes the three-task mean from the per-task values.

Simulation · Google Robot (RT-1 / Fractal) · SAPIEN (ManiSkill2)Report an error
89.0%Best · Xiaomi-Robotics-0
27Models ranked
Near saturationSaturation
Nov 18, 2026Numbers checked
#ModelTypeScoreDateSource
1Xiaomi-Robotics-0Xiaomi RoboticsVLA89.0%, self-reported by the authors, not a leaderboard entryFeb 2026
2X-VLATsinghua AIR et al.VLA88.3%, self-reported by the authors, not a leaderboard entryOct 2025
3MemoryVLATsinghua UniversityVLA87.8%, self-reported by the authors, not a leaderboard entryAug 2025
4RLDX-1RLWRLDVLA85.8%, self-reported by the authors, not a leaderboard entryMay 2026
5EO-1Shanghai AI LabVLA84.4%, self-reported by the authors, not a leaderboard entryAug 2025
6CogACTMicrosoft Research AsiaVLA82.7%, self-reported by the authors, not a leaderboard entryNov 2024
7RoboVLMs (KosMos P.H.)Tsinghua University & ByteDance ResearchVLA80.3%, self-reported by the authors, not a leaderboard entryDec 2024
8StarVLAStarVLA TeamVLA79.7%, self-reported by the authors, not a leaderboard entryApr 2026
9open-pi-zeroopen-pi-zero (Allen Z. Ren)VLA79.6%, self-reported by the authors, not a leaderboard entryJan 2, 2025
10HumeSJTU & Shanghai AI LabVLA78.7%, self-reported by the authors, not a leaderboard entryMay 2025
11villa-XMicrosoft Research AsiaVLA77.7%, self-reported by the authors, not a leaderboard entryJul 2025
12SpatialVLAShanghai AI LabVLA73.8%, self-reported by the authors, not a leaderboard entryJan 2025
13MolmoActAi2VLA71.6%, self-reported by the authors, not a leaderboard entryAug 2025
14VLA-JEPAUSTCVLA70.6%, self-reported by the authors, not a leaderboard entryFeb 2026
15DitaShanghai AI LabPolicy68.7%, self-reported by the authors, not a leaderboard entryMar 2025
16RT-1Google (Robotics at Google, Everyday Robots)VLA67.6%, reported by a third partyMay 2024
17π0.5Physical IntelligenceVLA67.5%, reported by a third partyJun 2026
18OpenVLA-OFTStanfordVLA63.0%, reported by a third partyJul 2025
19π0-FASTPhysical IntelligenceVLA61.9%, reported by a third partyAug 2025
20RT-2-XGoogle DeepMindVLA60.5%, reported by a third partyMay 2024
21π0Physical IntelligenceVLA58.8%, reported by a third partyAug 2025
22GR00T N1.5NVIDIAVLA57.9%, self-reported by the authors, not a leaderboard entrySep 26, 2025
23VITANanjing UniversityVLA57.4%, self-reported by the authors, not a leaderboard entryNov 2025
24RT-1-XGoogle DeepMindVLA49.4%, reported by a third partyMay 2024
25TraceVLAUniversity of MarylandVLA46.2%, self-reported by the authors, not a leaderboard entryDec 2024
26OpenVLAStanford / UC Berkeley et al.VLA32.7%, reported by a third partyJan 2025
27OctoUC Berkeley (RAIL)VLA14.6%, reported by a third partyMay 2024

Metric: Visual-matching success, mean of 3 Google Robot tasks (pick coke can, move near, open/close drawer). One result per model: RoboFYI re-run, then official leaderboard, then paper, then third party. Runs that deviate from the official protocol are listed on the model page but not ranked.

Wrong or missing number? Report an error