SimplerEnv WidowX leaderboard

Real-to-sim replica of the BridgeData V2 WidowX setup (spoon on towel, carrot on plate, stack cube, eggplant in basket). This WidowX protocol is reported separately from the Google Robot protocols.

Simulation · WidowX 250 (Bridge) · SAPIEN (ManiSkill2)Report an error
98.0%Best · StARe-VLA
32Models ranked
SaturatedSaturation
Nov 18, 2026Numbers checked
#ModelTypeScoreDateSource
1StARe-VLATU MunichVLA98.0%, self-reported by the authors, not a leaderboard entryDec 2025
2CORALFrontier RoboticsVLA97.9%, self-reported by the authors, not a leaderboard entryMar 2026
3SimVLAFrontier RoboticsVLA95.8%, self-reported by the authors, not a leaderboard entryFeb 2026
3X-VLATsinghua AIR et al.VLA95.8%, self-reported by the authors, not a leaderboard entryOct 2025
5Galaxea G0.5Galaxea AIVLA87.3%, self-reported by the authors, not a leaderboard entryAug 2026
6PDRVLAPeking University & JD.comVLA81.5%, self-reported by the authors, not a leaderboard entryOct 2026
7InternVLA-A1.5Shanghai AI LabVLA80.8%, self-reported by the authors, not a leaderboard entryJul 2026
8Xiaomi-Robotics-0Xiaomi RoboticsVLA79.2%, self-reported by the authors, not a leaderboard entryFeb 2026
9EO-1Shanghai AI LabVLA72.7%, self-reported by the authors, not a leaderboard entryAug 2025
10MemoryVLATsinghua UniversityVLA71.9%, self-reported by the authors, not a leaderboard entryAug 2025
10RLDX-1RLWRLDVLA71.9%, self-reported by the authors, not a leaderboard entryMay 2026
12InternVLA-M1Shanghai AI LabVLA71.7%, self-reported by the authors, not a leaderboard entryOct 2025
13VITANanjing UniversityVLA71.5%, self-reported by the authors, not a leaderboard entryNov 2025
14UniVLA (BAAI)CASIA & BAAIVLA69.8%, self-reported by the authors, not a leaderboard entryJun 2025
15open-pi-zeroopen-pi-zero (Allen Z. Ren)VLA68.4%, self-reported by the authors, not a leaderboard entryJan 2, 2025
16GR00T N1.5NVIDIAVLA67.8%, self-reported by the authors, not a leaderboard entrySep 26, 2025
17StarVLAStarVLA TeamVLA64.6%, self-reported by the authors, not a leaderboard entryApr 2026
18villa-XMicrosoft Research AsiaVLA62.5%, self-reported by the authors, not a leaderboard entryJul 2025
19HumeSJTU & Shanghai AI LabVLA60.8%, self-reported by the authors, not a leaderboard entryMay 2025
20F1-VLAShanghai AI LabVLA59.4%, self-reported by the authors, not a leaderboard entrySep 2025
21GR00T N1.7NVIDIAVLA59.3%, self-reported by the authors, not a leaderboard entryJun 12, 2026
22VLA-JEPAUSTCVLA57.3%, self-reported by the authors, not a leaderboard entryFeb 2026
23CogACTMicrosoft Research AsiaVLA51.3%, self-reported by the authors, not a leaderboard entryNov 2024
24GR00T N1.6NVIDIAVLA48.9%, self-reported by the authors, not a leaderboard entryJun 12, 2026
25UniVLA (OpenDriveLab)OpenDriveLabVLA47.9%, self-reported by the authors, not a leaderboard entryMay 2025
26SpatialVLAShanghai AI LabVLA42.7%, self-reported by the authors, not a leaderboard entryJan 2025
27MagmaMicrosoft ResearchVLA35.4%, self-reported by the authors, not a leaderboard entryFeb 2025
28π0Physical IntelligenceVLA27.8%, self-reported by the authors, not a leaderboard entryOct 2025
29OctoUC Berkeley (RAIL)VLA16.0%, self-reported by the authors, not a leaderboard entryMay 2024
30MoDEKITVLA7.3%, self-reported by the authors, not a leaderboard entryDec 2024
31RT-1-XGoogle DeepMindVLA1.1%, reported by a third partyMay 2024
32OpenVLAStanford / UC Berkeley et al.VLA1.0%, self-reported by the authors, not a leaderboard entryAug 2025

Metric: Visual-matching success, mean of 4 Bridge tasks. 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