| spatial-reasoning-on-viewspatial-bench | GPT-4o | – | Overall: 34.98Camera-based: 33.57Person-based: 36.29 |
| spatial-reasoning-on-viewspatial-bench | Gemini 2.0 Flash | – | Overall: 32.56Camera-based: 33.66Person-based: 31.53 |
| spatial-reasoning-on-viewspatial-bench | InternVL2.5-2B | – | Overall: 34.98Camera-based: 32.79Person-based: 37.04 |
| spatial-reasoning-on-viewspatial-bench | InternVL2.5-8B | – | Overall: 43.24Camera-based: 46.48Person-based: 40.20 |
| spatial-reasoning-on-viewspatial-bench | InternVL3-14B | – | Overall: 40.28Camera-based: 47.09Person-based: 33.88 |
| spatial-reasoning-on-viewspatial-bench | Kimi-VL-A3B-Instruct | – | Overall: 33.58Camera-based: 25.14Person-based: 41.52 |
| spatial-reasoning-on-viewspatial-bench | LLaVA-NeXT-Video-7B | – | Overall: 30.64Camera-based: 23.80Person-based: 37.07 |
| spatial-reasoning-on-viewspatial-bench | LLaVA-OneVision-7B | – | Overall: 27.49Camera-based: 28.49Person-based: 26.54 |
| spatial-reasoning-on-viewspatial-bench | Llama-3.2-11B-Vision | – | Overall: 28.82Camera-based: 23.73Person-based: 33.61 |
| spatial-reasoning-on-viewspatial-bench | MVSM-3B | – | Overall: 82.09Camera-based: 85.05Person-based: 79.31 |