3D Geometry on DL3DV (val)
11.9P-map ErrorWAN2.1-T2V-14B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| WAN2.1-T2V-14BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 11.9 | 4.4 | 0.614 | |
| WAN2.1-I2V-14BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 12.3 | 4.7 | 0.605 | |
| OpenSora2.0Probing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 14.1 | 6.6 | 0.56 | |
| WAN2.1-T2V-1.3BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 14.8 | 6.5 | 0.501 | |
| CogVideoX-T2V-2BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 16.5 | 8.9 | 0.492 | |
| AetherProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 16.5 | 8.4 | 0.495 | |
| CogVideoX-I2V-5BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 17.8 | 9.1 | 0.47 | |
| CogVideoX-T2V-5BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Video-Generation Models2026.05 | 17.9 | 9 | 0.475 | |
| Qwen3-VL-8BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 18 | 9.1 | 0.424 | |
| Qwen3-VL-4BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 19.1 | 9.3 | 0.39 | |
| InternVL3-8BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 20.2 | 10.6 | 0.362 | |
| Qwen3-VL-2BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 20.2 | 9.6 | 0.371 | |
| InternVL3.5-8BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 20.5 | 11 | 0.361 | |
| InternVL3.5-4BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 21.4 | 11.3 | 0.338 | |
| InternVL3-2BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 22.1 | 11.8 | 0.323 | |
| Qwen2.5-VL-7BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 23.4 | 12.7 | 0.31 | |
| InternVL3-1BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 26.2 | 12.5 | 0.254 | |
| Qwen2.5-VL-3BProbing=Frozen representations, Evaluation=Unified probe, Model Category=Vision-Language Models2026.05 | 32 | 15.2 | 0.163 |