Embodied Video Generation on PAI-Bench robot domain
93.26Domain ScorePF-Cosmos
Evaluation Results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PF-Cosmos2026.06 | 93.26 | — | — | — | — | — | — | — | — | 77.08 | 85.17 | — | — | — | — | — | |
| ABot-PhysWorldalignment=DPO2026.03 | 93.06 | 46.67 | 93.65 | 69.16 | 99.08 | 19.42 | 93.55 | 97.68 | 94.83 | 76.76 | 84.91 | — | — | — | — | — | |
| Abot-Physworld + DPOParam=14B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 93.06 | — | 93.7 | — | — | — | — | 97.7 | 94.8 | — | 84.91 | 46.7 | 69.2 | 99.1 | 93.6 | 19.4 | |
| Abot-PhysWorld2026.06 | 93.06 | — | — | — | — | — | — | — | — | 76.76 | 84.91 | — | — | — | — | — | |
| Cosmos3-Nanovanilla finetuning=true2026.06 | 91.54 | — | — | — | — | — | — | — | — | 76.52 | 84.03 | — | — | — | — | — | |
| KairosParam=4B, Model Scale Group=Small-scale Models (<10B)2026.06 | 88.59 | — | 93.2 | — | — | — | — | 97.8 | 94.8 | — | 82.57 | 46.8 | 69 | 99.1 | 93.6 | 18.1 | |
| PF-Wan2026.06 | 88.2 | — | — | — | — | — | — | — | — | 76.26 | 81.73 | — | — | — | — | — | |
| Cosmos3-NanoParam=16B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 88.04 | — | 92.1 | — | — | — | — | 98.2 | 95.3 | — | 82.62 | 48.4 | 73 | 99.3 | 91.7 | 19.7 | |
| ABot-PhysWorldalignment=SFT2026.03 | 87.85 | 46.2 | 93.73 | 69.06 | 99.16 | 19.27 | 94.06 | 97.77 | 94.98 | 76.78 | 82.32 | — | — | — | — | — | |
| Abot-PhysworldParam=14B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 87.85 | — | 93.7 | — | — | — | — | 97.8 | 95 | — | 82.32 | 46.2 | 69.1 | 99.2 | 94.1 | 19.3 | |
| Wan 2.52026.03 | 86.44 | 54.77 | 89.85 | 64.58 | 96.23 | 21.9 | 87.84 | 94.38 | 94.28 | 75.48 | 80.96 | — | — | — | — | — | |
| Wan2.5Param=/, Model Scale Group=Large-scale Models (≥10B)2026.06 | 86.44 | — | 89.9 | — | — | — | — | 94.4 | 94.3 | — | 80.96 | 54.8 | 64.6 | 96.2 | 87.8 | 21.9 | |
| Wan2.52026.06 | 86.44 | — | — | — | — | — | — | — | — | 75.48 | 80.96 | — | — | — | — | — | |
| GigaWorld-02026.03 | 85.83 | 47.57 | 92.19 | 65.06 | 99.08 | 19.44 | 91.11 | 96.73 | 96.07 | 75.91 | 80.87 | — | — | — | — | — | |
| GigaWorld-0Param=2B, Model Scale Group=Small-scale Models (<10B)2026.06 | 85.83 | — | 92.2 | — | — | — | — | 96.7 | 96.1 | — | 80.87 | 47.6 | 65.1 | 99.1 | 91.1 | 19.4 | |
| GigaWorld-02026.06 | 85.83 | — | — | — | — | — | — | — | — | 75.91 | 80.87 | — | — | — | — | — | |
| Wan2.2-I2V-A14Bvanilla finetuning=true2026.06 | 84.42 | — | — | — | — | — | — | — | — | 75.38 | 79.9 | — | — | — | — | — | |
| Wan2.1_14B2026.03 | 83.91 | 47.23 | 93.15 | 71.18 | 99.17 | 19.21 | 91.85 | 97.45 | 94.51 | 76.72 | 80.32 | — | — | — | — | — | |
| Wan2.1Param=14B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 83.91 | — | 93.2 | — | — | — | — | 97.5 | 94.5 | — | 80.32 | 47.2 | 71.2 | 99.2 | 91.9 | 19.2 | |
| Veo 3.12026.03 | 83.5 | 54.58 | 92.16 | 72.44 | 97.12 | 22.14 | 91.46 | 93.17 | 96.14 | 77.4 | 80.45 | — | — | — | — | — | |
| Veo 3.1Param=/, Model Scale Group=Large-scale Models (≥10B)2026.06 | 83.5 | — | 92.2 | — | — | — | — | 93.2 | 96.1 | — | 80.45 | 54.6 | 72.4 | 97.1 | 91.5 | 22.1 | |
| Veo 3.12026.06 | 83.5 | — | — | — | — | — | — | — | — | 77.4 | 80.45 | — | — | — | — | — | |
| WoW-wan 14B2026.03 | 83.01 | 46.64 | 92.95 | 70.27 | 98.58 | 19.41 | 91.49 | 96.13 | 92.92 | 76.05 | 79.53 | — | — | — | — | — | |
| WoW-wanParam=14B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 83.01 | — | 93 | — | — | — | — | 96.1 | 92.9 | — | 79.53 | 46.6 | 70.3 | 98.6 | 91.5 | 19.4 | |
| WoW-Wan 14B2026.06 | 83.01 | — | — | — | — | — | — | — | — | 76.05 | 79.53 | — | — | — | — | — | |
| LingbotParam=28B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 82.98 | — | 93.1 | — | — | — | — | 97.7 | 97.6 | — | 79.97 | 50.2 | 67.1 | 98.9 | 91.2 | 19.8 | |
| Cosmos-Predict2.5Param=14B, Model Scale Group=Large-scale Models (≥10B)2026.06 | 82.6 | — | 93.1 | — | — | — | — | 94.3 | 92.2 | — | 79.4 | 48 | 72 | 99.1 | 91.3 | 19.2 | |
| Wan2.2-I2V-A14B2026.06 | 81.7 | — | — | — | — | — | — | — | — | 76.15 | 78.93 | — | — | — | — | — | |
| Cosmos-Predict2.5Param=2B, Model Scale Group=Small-scale Models (<10B)2026.06 | 80.44 | — | 92 | — | — | — | — | 93.7 | 91.2 | — | 78.26 | 49.3 | 74.1 | 99.1 | 90.2 | 19.1 | |
| Cosmos-Predict 2.52026.03 | 80.21 | 48.97 | 91.66 | 74.05 | 99.06 | 19.11 | 89.73 | 93.04 | 90.3 | 75.74 | 77.97 | — | — | — | — | — | |
| Wan2.2Param=5B, Model Scale Group=Small-scale Models (<10B)2026.06 | 80.17 | — | 92.3 | — | — | — | — | 97.8 | 97.4 | — | 78.63 | 49.3 | 70.6 | 99.1 | 90.8 | 19.4 | |
| Sora v2 Pro2026.03 | 76.26 | 53.24 | 92.85 | 69.56 | 97.02 | 22.03 | 91.63 | 95.07 | 92.9 | 76.79 | 76.52 | — | — | — | — | — | |
| Sora v2 ProParam=/, Model Scale Group=Large-scale Models (≥10B)2026.06 | 76.26 | — | 92.9 | — | — | — | — | 95.1 | 92.9 | — | 76.52 | 53.2 | 69.6 | 97 | 91.6 | 22 | |
| Sora v2 Pro2026.06 | 76.26 | — | — | — | — | — | — | — | — | 76.79 | 76.52 | — | — | — | — | — | |
| UnifoLM-WMA-02026.03 | 66.93 | 45.47 | 94.23 | 65.64 | 98.75 | 18.78 | 94.12 | 96.38 | 94.03 | 75.93 | 71.43 | — | — | — | — | — | |
| UnifoLM-WMA-0Param=3B, Model Scale Group=Small-scale Models (<10B)2026.06 | 66.93 | — | 94.2 | — | — | — | — | 96.4 | 94 | — | 71.43 | 45.5 | 65.6 | 98.8 | 94.1 | 18.8 |