Robotic Manipulation on SIMPLER Google Robot VA
98Pick Up Coke Can Success RateDAM-VLA
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| DAM-VLA2026.03 | 98 | 74 | 68 | 84 | 81 | — | — | |
| CogACT2026.03 | 96 | 84 | 29 | 40 | 62 | — | — | |
| LatBot2025.11 | 95.7 | 78.3 | 73 | — | 70.1 | 33.3 | — | |
| EfficientVLARetained tokens=562026.03 | 94.4 | 77.2 | 27.6 | — | 62.6 | 51.3 | — | |
| VLA-CacheRetained tokens=562026.03 | 91.7 | 79.3 | 32.5 | — | 62.3 | 45.8 | — | |
| FastVRetained tokens=562026.03 | 91.4 | 78.6 | 27.6 | — | 62.1 | 50.6 | — | |
| RT-12026.03 | 90 | 46 | 35 | 3 | 44 | — | — | |
| Hard-TIESRetained tokens=562026.03 | 89.7 | 74.7 | 41 | — | 65 | 54.6 | — | |
| CogACTPrecision=FP Model2026.02 | 89.6 | 80.8 | 28.3 | — | 61.3 | 46.6 | — | |
| Soft-TIESRetained tokens=562026.03 | 88.7 | 76.2 | 44.3 | — | 67.6 | 61.1 | — | |
| CogACTRetained tokens=562026.03 | 88.6 | 76.5 | 28.8 | — | 59.9 | 45.6 | — | |
| SpatialVLA2025.11 | 88 | 72.7 | 41.8 | — | 51.8 | 6.3 | — | |
| SpatialVLA2026.03 | 88 | 73 | 42 | — | — | — | — | |
| π02025.11 | 85.2 | 40.8 | 42.1 | — | 46 | 16 | — | |
| HBVLAWeight Bit-width=1.08 bit2026.02 | 85.1 | 72.5 | 21.4 | — | 51 | 24.9 | -10.3 | |
| DD-VLA2025.11 | 82.5 | 64.6 | 23.6 | — | — | — | — | |
| RT-2-X2025.11 | 82.3 | 79.2 | 35.3 | — | 54.4 | 20.6 | — | |
| RT-2-X2026.03 | 82 | 79 | 35 | 21 | 54 | — | — | |
| MemoryVLA2025.11 | 80.5 | 78.8 | 53.2 | — | 67.7 | 58.3 | — | |
| HBLLMWeight Bit-width=1.08 bit2026.02 | 79.3 | 69.8 | 24.6 | — | 47.9 | 18 | -13.4 | |
| RoboVLM2026.03 | 76 | 60 | 11 | — | — | — | — | |
| BiVLMWeight Bit-width=1.08 bit2026.02 | 71.8 | 67.3 | 23.1 | — | 47.4 | 27.5 | -13.9 | |
| OpenVLA2026.03 | 64 | 64 | 19 | 1 | 37 | — | — | |
| OpenVLA2025.11 | 60.8 | 67.7 | 28.3 | — | 39.3 | 1.2 | — | |
| RT-1-X2026.03 | 49 | 33 | 29 | 10 | 30 | — | — | |
| BiLLMWeight Bit-width=1.08 bit2026.02 | 21.1 | 22.2 | 15.3 | — | 14.9 | 0.8 | -46.4 | |
| Octo-Base2026.03 | 1 | 4 | 1 | 0 | 1 | — | — | |
| VideoVLA2025.12 | 0.898 | 0.733 | 0.288 | 0.593 | 0.628 | — | — | |
| CogACT2025.12 | 0.896 | 0.808 | 0.283 | 0.471 | 0.614 | — | — | |
| SpatialVLA2025.12 | 0.895 | 0.717 | 0.362 | 0.122 | 0.524 | — | — | |
| RT-2-X2025.12 | 0.823 | 0.792 | 0.353 | 0.206 | 0.544 | — | — | |
| π0Input=visual observations only2025.12 | 0.714 | 0.579 | 0.265 | 0.18 | 0.434 | — | — | |
| OpenVLA2025.12 | 0.608 | 0.677 | 0.288 | 0.005 | 0.394 | — | — | |
| RT-1-X2025.12 | 0.49 | 0.323 | 0.294 | 0.111 | 0.305 | — | — | |
| Octo-Base2025.12 | 0.006 | 0.031 | 0.011 | 0 | 0.012 | — | — |