Robotics Task Execution on RoboTwin 2.0 (Clean)
93.8Success RateWAM4D
Evaluation Results
| Method | Links | |
|---|---|---|
| WAM4DLatency (ms)=525.43 ± 5.64, VRAM (GiB)=9.71, Action-denoising steps=10, Video-denoising steps=52026.06 | 93.8 | |
| LingBot-VALatency (ms)=843.57 ± 11.55, VRAM (GiB)=12.97, Action-denoising steps=10, Video-denoising steps=52026.06 | 92.9 | |
| Fast-WAMLatency (ms)=425.53 ± 6.01, VRAM (GiB)=11.55, Action-denoising steps=102026.06 | 91.9 | |
| HyVLA-0.52026.06 | 90.9 | |
| JoyAI-RA2026.06 | 90.5 | |
| HyVLA-0.5compact memory encoder=false2026.06 | 88.8 | |
| MotusLatency (ms)=1516.30 ± 10.64, VRAM (GiB)=11.55, Action-denoising steps=102026.06 | 88.7 | |
| Motus2026.06 | 88.7 | |
| starVLA2026.06 | 88.2 | |
| HyVLA-0.5compact memory encoder=false, UMI pre-training=false2026.06 | 88.1 | |
| LingBot-VLA2026.06 | 86.5 | |
| Qwen-VLA2026.06 | 86.1 | |
| ImageWAMP.T. (Policy Pre-training)=false, Keyframe prediction (K.F.)=false2026.06 | 84.4 | |
| π0.5Latency (ms)=72.03 ± 0.06, VRAM (GiB)=8.45, Action-denoising steps=102026.06 | 82.7 | |
| π0.52026.06 | 82.7 | |
| ABot-M02026.06 | 81.2 | |
| BagelVLA (w/ K.F.)P.T. (Policy Pre-training)=true, Keyframe prediction (K.F.)=true2026.06 | 75.3 | |
| π0Latency (ms)=64.16 ± 0.06, VRAM (GiB)=8.45, Action-denoising steps=102026.06 | 65.9 | |
| π02026.06 | 65.9 | |
| BagelVLA (w/o K.F.)P.T. (Policy Pre-training)=true, Keyframe prediction (K.F.)=false2026.06 | 56.7 |