Robot Manipulation on RoboTwin 2.0
97.4Average Success RateLingBot-VA
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LingBot-VAPolicy Category=WAM-Based, Reference Note=Different RoboTwin training settings2026.06 | 97.4 | — | 98 | — | — | — | — | 96 | 99 | 97 | — | — | — | — | — | — | 99 | 93 | — | — | — | — | — | — | — | — | — | — | — | — | 97 | — | — | — | — | — | — | — | — | — | — | — | — | — | 100 | — | — | |
| Fast-WAMPolicy Category=WAM-Based, Reference Note=Different RoboTwin training settings2026.06 | 96.6 | — | 97 | — | — | — | — | 100 | 100 | 88 | — | — | — | — | — | — | 96 | 93 | — | — | — | — | — | — | — | — | — | — | — | — | 99 | — | — | — | — | — | — | — | — | — | — | — | — | — | 100 | — | — | |
| LingBot-VAType=WAM, Params=5.3B, EPT=w/2026.06 | 92.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92.9 | 91.5 | — | — | — | |
| dVLA-RLPolicy Category=VLA-Based2026.06 | 92 | — | 95.3 | — | — | — | — | 96.9 | 89.1 | 87.5 | — | — | — | — | — | — | 95.3 | 79.7 | — | — | — | — | — | — | — | — | — | — | — | — | 96.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | 95.3 | — | — | |
| Fast-WAMType=WAM, Params=6B, EPT=w/o2026.06 | 91.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 91.9 | 91.8 | — | — | — | |
| MotusType=WAM, Params=8B, EPT=w/2026.06 | 87.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 88.7 | 87 | — | — | — | |
| MotusPolicy Category=WAM-Based, Reference Note=Different RoboTwin training settings2026.06 | 84.6 | — | 88 | — | — | — | — | 63 | 99 | 74 | — | — | — | — | — | — | 90 | 79 | — | — | — | — | — | — | — | — | — | — | — | — | 86 | — | — | — | — | — | — | — | — | — | — | — | — | — | 98 | — | — | |
| GigaWorldPolicy Category=WAM-Based, Reference Note=Different RoboTwin training settings2026.06 | 83.8 | — | 86 | — | — | — | — | 72 | 98 | 78 | — | — | — | — | — | — | 86 | 88 | — | — | — | — | — | — | — | — | — | — | — | — | 72 | — | — | — | — | — | — | — | — | — | — | — | — | — | 90 | — | — | |
| DP3+COREConditioning=Visual CORE, Observation=Point-cloud, NFE=102026.06 | 82.3 | 98.8 | 89.2 | — | 87 | — | 90.6 | — | — | 91.4 | — | — | — | 80.6 | 92.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 32 | 72 | 88.8 | — | — | — | — | — | — | — | |
| LingBot-VAType=WAM, Params=5.3B, EPT=w/o2026.06 | 80.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 80.6 | — | — | — | — | |
| π0.5Type=VLA, Params=3B, EPT=w/2026.06 | 79.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 82.7 | 76.8 | — | — | — | |
| MP1+COREConditioning=Visual CORE, Observation=Point-cloud, NFE=12026.06 | 79.4 | 100 | 91.6 | — | 90.4 | — | 90 | — | — | 87.6 | — | — | — | 89 | 55.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 33.6 | 69.6 | 86.8 | — | — | — | — | — | — | — | |
| DP3Observation=Point-cloud, NFE=102026.06 | 77.5 | 98.6 | 80.2 | — | 85 | — | 88.4 | — | — | 84 | — | — | — | 73.4 | 79.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 30.6 | 69.2 | 86.4 | — | — | — | — | — | — | — | |
| DP3+TEXTConditioning=Text, Observation=Point-cloud, NFE=102026.06 | 77.4 | 96.6 | 79.8 | — | 86.2 | — | 88.8 | — | — | 84.6 | — | — | — | 73.8 | 78.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 30.2 | 69.6 | 86 | — | — | — | — | — | — | — | |
| Light-WAMType=WAM, Params=2B, EPT=w/o2026.06 | 76.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 76.4 | 76.3 | — | — | — | |
| MotusType=WAM, Params=8B, EPT=w/o2026.06 | 74.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.8 | 77 | — | — | — | |
| π0.5Policy Category=VLA-Based2026.06 | 74.9 | — | 88 | — | — | — | — | 32 | 88 | 72 | — | — | — | — | — | — | 98 | 74 | — | — | — | — | — | — | — | — | — | — | — | — | 73 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74 | — | — | |
| X-VLAType=VLA, Params=0.9B, EPT=w/2026.06 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.9 | 72.8 | — | — | — | |
| MP1+TEXTConditioning=Text, Observation=Point-cloud, NFE=12026.06 | 69.7 | 96.8 | 82.6 | — | 72.2 | — | 55.6 | — | — | 83.8 | — | — | — | 80.4 | 44.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 31.4 | 63.2 | 85.8 | — | — | — | — | — | — | — | |
| SimpleVLA-RLPolicy Category=VLA-Based2026.06 | 68.7 | — | 87.5 | — | — | — | — | 89.2 | 64.1 | 61.2 | — | — | — | — | — | — | 68.3 | 45.3 | — | — | — | — | — | — | — | — | — | — | — | — | 39.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | 94.2 | — | — | |
| MP1Observation=Point-cloud, NFE=12026.06 | 68.3 | 96.6 | 83.4 | — | 63.8 | — | 56.8 | — | — | 83.4 | — | — | — | 80.4 | 40.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 31 | 62.4 | 85 | — | — | — | — | — | — | — | |
| CAPSNumber of candidate actions=50, Number of seeds=100, Number of scenes=1002026.05 | 66.2 | 95 | 81 | 46 | 89 | 94 | 38 | 72 | 76 | 61 | 53 | 72 | 21 | 74 | 27 | 62 | 71 | 54 | 47 | 63 | 69 | 93 | 79 | 87 | 33 | 36 | 25 | 76 | 57 | 81 | 89 | 67 | 58 | 69 | 100 | 98 | 52 | 89 | 74 | 97 | 42 | 47 | — | — | — | — | — | |
| CFP (Unet)Architecture=Unet2026.06 | 66 | — | 77 | — | — | — | — | 88 | — | — | — | 39 | — | — | — | 51 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 59 | 81 | 68 | |
| π0.5 + TACONumber of candidate actions=50, Number of seeds=100, Number of scenes=1002026.05 | 64 | 93 | 79 | 40 | 87 | 98 | 36 | 63 | 66 | 57 | 54 | 88 | 18 | 67 | 21 | 59 | 73 | 56 | 42 | 65 | 61 | 92 | 72 | 87 | 28 | 32 | 24 | 78 | 52 | 78 | 86 | 65 | 56 | 67 | 100 | 99 | 45 | 91 | 68 | 93 | 38 | 49 | — | — | — | — | — | |
| CFP (DiT)Architecture=DiT2026.06 | 63 | — | 61 | — | — | — | — | 82 | — | — | — | 41 | — | — | — | 28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 42 | 93 | 79 | |
| π0Type=VLA, Params=3B, EPT=w/2026.06 | 62.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 65.9 | 58.4 | — | — | — | |
| MM-ACT*Policy Category=VLA-Based2026.06 | 61.4 | — | 85 | — | — | — | — | 49 | 46 | 40 | — | — | — | — | — | — | 71 | 57 | — | — | — | — | — | — | — | — | — | — | — | — | 62 | — | — | — | — | — | — | — | — | — | — | — | — | — | 81 | — | — | |
| π0.5Number of candidate actions=50, Number of seeds=100, Number of scenes=1002026.05 | 59.3 | 89 | 69 | 36 | 86 | 100 | 24 | 58 | 62 | 42 | 55 | 85 | 13 | 67 | 20 | 52 | 72 | 51 | 39 | 62 | 62 | 89 | 68 | 85 | 23 | 34 | 16 | 69 | 45 | 68 | 76 | 53 | 54 | 62 | 100 | 98 | 43 | 81 | 64 | 93 | 26 | 42 | — | — | — | — | — | |
| MBA2026.06 | 56 | — | 61 | — | — | — | — | 71 | — | — | — | 37 | — | — | — | 27 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 39 | 84 | 71 | |
| FlowPolicyNFE=12026.06 | 54.7 | 93.6 | 56.8 | — | 79.4 | — | 38.6 | — | — | 48.2 | — | — | — | 71.6 | 21.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 18.8 | 49.6 | 69 | — | — | — | — | — | — | — | |
| π0Policy Category=VLA-Based2026.06 | 52.1 | — | 59 | — | — | — | — | 96 | 51 | 41 | — | — | — | — | — | — | 50 | 38 | — | — | — | — | — | — | — | — | — | — | — | — | 22 | — | — | — | — | — | — | — | — | — | — | — | — | — | 60 | — | — | |
| 3D-FDP2026.06 | 47 | — | 58 | — | — | — | — | 65 | — | — | — | 41 | — | — | — | 16 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 27 | 80 | 44 | |
| DP32026.06 | 43 | — | 39 | — | — | — | — | 64 | — | — | — | 32 | — | — | — | 18 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 27 | 66 | 54 | |
| DPObservation=Image-based, NFE=102026.06 | 42.4 | 90.6 | 48.2 | — | 50.8 | — | 12 | — | — | 38.2 | — | — | — | 46.6 | 8.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 6 | 62.8 | 61 | — | — | — | — | — | — | — | |
| RDTPolicy Category=VLA-Based2026.06 | 36.1 | — | 22 | — | — | — | — | 95 | 45 | 33 | — | — | — | — | — | — | 18 | 21 | — | — | — | — | — | — | — | — | — | — | — | — | 13 | — | — | — | — | — | — | — | — | — | — | — | — | — | 42 | — | — | |
| RDTNumber of candidate actions=50, Number of seeds=100, Number of scenes=1002026.05 | 34.6 | 81 | 77 | 0 | 64 | 74 | 45 | 90 | 72 | 25 | 8 | 43 | 2 | 59 | 37 | 2 | 42 | 3 | 1 | 10 | 5 | 50 | 6 | 78 | 4 | 12 | 1 | 33 | 1 | 15 | 15 | 35 | 33 | 50 | 84 | 74 | 2 | 21 | 51 | 76 | 1 | 35 | — | — | — | — | — | |
| OpenVLA-OFTPolicy Category=VLA-Based2026.06 | 34.2 | — | 28.1 | — | — | — | — | 45.3 | 10.1 | 28.1 | — | — | — | — | — | — | 29.7 | 37.5 | — | — | — | — | — | — | — | — | — | — | — | — | 17.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.3 | — | — | |
| 2D AblationCross Embodiments=True2026.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 64.8 | 35.2 | — | — | — | |
| Dex-BEVCross Embodiments=True2026.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 76 | 42 | — | — | — | |
| X-VLACross Embodiments=False2026.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 70 | 39 | — | — | — | |
| π0Cross Embodiments=False2026.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 46.4 | 16.4 | — | — | — |