Robotic Manipulation on SIMPLER Google Robot Visual Matching
97.2PickCan Success RateAC2-VLA
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| AC2-VLABackbone=Prismatic-7B, Action Head=DiT-Base, Action Horizon=15, Diffusion Steps=8, Action Distillation=true, Max Token Pruning Ratio=0.6, Cache Reuse Threshold=0.22026.01 | 97.2 | 82.7 | 80.6 | 46.8 | 76.8 | |
| LatBot2025.11 | 96.7 | 91.7 | 90.4 | 33.3 | 78 | |
| PriGo-DPEvaluation Setting=Visual Matching, Base Model=CogACT2026.07 | 93.2 | 89.3 | 77.5 | 59.7 | 79.9 | |
| VideoVLA2025.12 | 92.3 | 82.9 | 66.2 | 50.9 | 73.1 | |
| CogACTPrecision=Full precision2026.01 | 91.3 | 85 | 71.8 | 50.9 | 74.8 | |
| CogACT2025.12 | 91.3 | 85 | 71.8 | 52.7 | 75.2 | |
| CogACTEvaluation Setting=Visual Matching2026.07 | 91.3 | 85 | 71.8 | 50.9 | 74.8 | |
| MemoryVLA2025.11 | 90.7 | 88 | 84.7 | 47.2 | 77.2 | |
| π02025.11 | 87.3 | 35 | 72.6 | 16 | 52.7 | |
| SpatialVLA2025.11 | 86 | 77.9 | 57.4 | 0 | 55.3 | |
| RT-1Precision=Full precision2026.01 | 85.7 | 44.2 | 73 | 6.5 | 52.4 | |
| DD-VLA2025.11 | 85.4 | 67.5 | 60.6 | — | — | |
| π0Input=visual observations only2025.12 | 84 | 55.8 | 34.3 | 31.5 | 53.5 | |
| villa-X2025.11 | 81.7 | 55.4 | 38.4 | — | — | |
| SpatialVLA2025.12 | 81 | 69.6 | 59.3 | 8.3 | 54.6 | |
| RT-2-XPrecision=Full precision2026.01 | 78.7 | 77.9 | 25 | 3.7 | 46.3 | |
| RT-2-X2025.12 | 78.7 | 77.9 | 25 | 3.7 | 46.3 | |
| RT-2-X2025.11 | 78.7 | 77.9 | 25 | 3.7 | 46.3 | |
| RT-2-XEvaluation Setting=Visual Matching2026.07 | 78.7 | 77.9 | 25 | 3.7 | 46.3 | |
| RoboVLM2025.11 | 76.3 | 79 | 44.9 | 27.8 | 57 | |
| RT-1-XPrecision=Full precision2026.01 | 56.7 | 31.7 | 59.7 | 21.3 | 42.4 | |
| RT-1-X2025.12 | 56.7 | 31.7 | 59.7 | 22.2 | 42.7 | |
| OpenVLAPrecision=Full precision2026.01 | 18 | 56.3 | 63 | 0 | 34.3 | |
| OpenVLA2025.12 | 18 | 56.3 | 63 | 0 | 34.3 | |
| OpenVLA2025.11 | 18 | 56.3 | 63 | 0 | 34.3 | |
| OpenVLAEvaluation Setting=Visual Matching2026.07 | 18 | 56.3 | 63 | 0 | 34.3 | |
| Octo-BasePrecision=Full precision2026.01 | 17 | 4.2 | 22.7 | 0 | 11 | |
| Octo-Base2025.12 | 17 | 4.2 | 22.7 | 0 | 11 | |
| Octo-BaseEvaluation Setting=Visual Matching2026.07 | 17 | 4.2 | 22.7 | 0 | 11 |