Multimodal Reasoning on V*
89.5Pass@1Seed 1.5-VL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Seed 1.5-VLthinking=false, decoding=greedy2025.05 | 89.5 | — | |
| Seed 1.5-VLthinking=true, decoding=greedy2025.05 | 89 | — | |
| Claude 3.7 Sonnetthinking=true, decoding=sampling2025.05 | 86.4 | — | |
| Qwen 2.5-VL 72Bthinking=false, decoding=greedy2025.05 | 86.4 | — | |
| Gemini 1.5 Prothinking=true, decoding=greedy2025.05 | 79.1 | — | |
| GPT-4othinking=false, decoding=greedy2025.05 | 73.9 | — | |
| OpenAI o1thinking=true, decoding=greedy2025.05 | 69.7 | — | |
| DAM-3Bmask=12025.10 | — | 45 | |
| GAR-8Bmask=12025.10 | — | 59.2 | |
| GAR-8Bmask=1, Training Data=w/ 600K General Data2025.10 | — | 62.3 | |
| PAM-3Bmask=12025.10 | — | 1.4 | |
| PerceptionLM-8Bmask=12025.10 | — | 69.1 |