| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| LWilder | Perplexity3.8 | 28 | 5mo ago | ||
| LiveB | PPL5.1 | 28 | 5mo ago | ||
| WeMath | ZPPO | Accuracy82.6 | 16 | 1mo ago | |
| VBlind | ZPPO | Accuracy60.5 | 16 | 1mo ago | |
| VisP | ZPPO | Accuracy42 | 16 | 1mo ago | |
| OCRZH | ZPPO | Accuracy26 | 16 | 1mo ago | |
| OCREN | ZPPO | Accuracy48.8 | 16 | 1mo ago | |
| MMMUPro | ZPPO | Accuracy53.2 | 16 | 1mo ago | |
| MVision | ZPPO | Accuracy50.5 | 16 | 1mo ago | |
| MVerse | ZPPO | Accuracy76 | 16 | 1mo ago | |
| InfoVQA | ZPPO | Accuracy84.6 | 16 | 1mo ago | |
| EmbSp | ZPPO | Accuracy79.5 | 16 | 1mo ago | |
| DynaM | ZPPO | Accuracy52.7 | 16 | 1mo ago | |
| CharXiv | ZPPO | Accuracy73.9 | 16 | 1mo ago | |
| BabyV | ZPPO | Accuracy18.6 | 16 | 1mo ago | |
| AI2D | ZPPO | Accuracy85.3 | 16 | 1mo ago | |
| 16 VLM benchmark suite macro-average | PASS@1 Accuracy75.5 | 9 | 1mo ago | ||
| LCOCO | PPL4.51 | 7 | 5mo ago | ||
| LiveB, LWilder, and LCOCO Combined | CASP_GPTQ | Avg Relative Improvement20.1 | 3 | 5mo ago |
