| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| BC-VL | GPT-5 | Accuracy57.6 | 37 | 22d ago | |
| HLE-VL | Gemini-2.5 Pro | Accuracy (HLE-VL)19 | 18 | 22d ago | |
| MMSrch | Accuracy82.94 | 18 | 1mo ago | ||
| FVQA | Accuracy76.67 | 16 | 1mo ago | ||
| MMBC | Gemini-2.5 Pro | Accuracy13.8 | 15 | 22d ago | |
| HR-MMSrch | SenseNova-MARS-32B | Accuracy54.43 | 15 | 1mo ago | |
| MTA (test) | MTA-DeepSearch-32B | Accuracy29.78 | 12 | 1mo ago | |
| MMSrch-Plus | Accuracy33.51 | 12 | 1mo ago |