Temporal spatial reasoning on VSTI-Bench (test)
68.9Average ScoreCambrian-P
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Cambrian-PPose Supervision=true2026.05 | 68.9 | 42.5 | 46.6 | 87.7 | 94.3 | 73.2 | |
| GeoThinkerParameters=8B2026.05 | 67.4 | 38.4 | 45.8 | 84.2 | 93.6 | 75.2 | |
| Cambrian-PPose Supervision=false2026.05 | 62.4 | 39.4 | 40.6 | 67.7 | 92.2 | 72 | |
| VLM-3RParameters=7B2026.05 | 58.8 | 39.4 | 39.6 | 60.6 | 86.5 | 68.6 | |
| SpatialMosaicVLMRank=1, Category=Open-sourced VLMs, Backbone=7B, Training Data=SpatialMosaic, Protocol=Zero-shot2025.12 | 46.8 | 27.8 | 5 | 48.2 | 74.2 | 52.3 | |
| VLM-3RRank=2, Category=Open-sourced VLMs, Backbone=7B, Training Data=SpatialMosaic, Protocol=Zero-shot2025.12 | 46.7 | 28.6 | 7.1 | 47.9 | 73.9 | 53 | |
| LLaVA-NeXT-Video-72BRank=3, Category=Open-sourced VLMs2025.12 | 44 | 32.3 | 10.5 | 48.1 | 78.3 | 50.9 | |
| LLaVA-NeXT-VideoParameters=72B2026.05 | 44 | 32.3 | 10.5 | 48.1 | 78.3 | 50.9 | |
| InternVL2-8BRank=4, Category=Open-sourced VLMs2025.12 | 43.5 | 32.9 | 13.5 | 48 | 68 | 55 | |
| LLaVA-OneVision-7BRank=5, Category=Open-sourced VLMs2025.12 | 41.7 | 29.9 | 19.3 | 47.5 | 62.1 | 49.8 | |
| LLaVA-NeXT-Video-7BRank=6, Category=Open-sourced VLMs2025.12 | 40 | 28.2 | 1.8 | 49.8 | 64.7 | 55.6 | |
| GPT-4oRank=7, Category=Proprietary Models (API)2025.12 | 38.2 | 29.5 | 23.4 | 37.3 | 58.1 | 42.5 | |
| VILA-1.5-40BRank=7, Category=Open-sourced VLMs2025.12 | 38.2 | 28.2 | 15.7 | 28.8 | 65.4 | 53 | |
| GPT-4o2026.05 | 38.2 | 29.5 | 23.4 | 37.3 | 58.1 | 42.5 | |
| InternVL2-2BRank=9, Category=Open-sourced VLMs2025.12 | 38.1 | 17.7 | 27.8 | 43 | 54.9 | 47.2 | |
| VILA-1.5-8BRank=10, Category=Open-sourced VLMs2025.12 | 37.3 | 30.1 | 27.3 | 42.2 | 50.4 | 36.7 | |
| LLaVA-OneVision-0.5BRank=11, Category=Open-sourced VLMs2025.12 | 36.9 | 16.5 | 32.4 | 46.1 | 50.5 | 39 | |
| LongVA-7BRank=12, Category=Open-sourced VLMs2025.12 | 32.3 | 13.5 | 5.1 | 43.7 | 57.9 | 41.2 | |
| Gemini-1.5 FlashRank=13, Category=Proprietary Models (API)2025.12 | 32.1 | 28.5 | 20.9 | 24.4 | 52.6 | 33.9 | |
| Gemini-1.5 Flash2026.05 | 32.1 | 28.5 | 20.9 | 24.4 | 52.6 | 33.9 |