| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| DocVQA (test) | Qwen2.5-VL-72B | Accuracy96.4 | 78 | 18d ago | |
| DocVQA | ANLS97.87 | 52 | 1mo ago | ||
| Qasper | FP8 | Accuracy40.8 | 44 | 1mo ago | |
| LongDocURL | GPT-4o-200b-128 | Accuracy (All)71.4 | 30 | 5d ago | |
| DUDE | GPT-4o-200b-128 | Accuracy (all)80.7 | 23 | 1mo ago | |
| MMLongBench-Doc | GPT-4o-200b-128 | Accuracy (all)69.6 | 23 | 1mo ago | |
| SlideVQA (test) | Eagle-2.5 | EM63.2 | 19 | 1mo ago | |
| PathSpatial-DocQA | LAMMI | ACS80.9 | 13 | 1mo ago | |
| DocVQA | MODIX | EM (Exact Match)91.02 | 12 | 3d ago | |
| DUDE | Arctic-TILT | ANLS0.5809 | 12 | 1mo ago | |
| InfoVQA | MiMoVL 7B-RL | Score90.1 | 11 | 11d ago | |
| M3DocVQA | BookRAG | Exact Match61 | 11 | 1mo ago | |
| MMLongBench | BookRAG | Exact Match43.8 | 11 | 1mo ago | |
| DocBench | DocDancer | LasJ Score85.5 | 10 | 1mo ago | |
| PubMed Sci-papers | Qwen3-8B | Accuracy55.07 | 9 | 24d ago | |
| Arxiv Sci-papers | Qwen3-8B | Accuracy54.26 | 9 | 24d ago | |
| InfoVQA (val) | Score76.12 | 7 | 1mo ago | ||
| MPVQA (test-server) | MultiDocFusion | ANLS0.1544 | 6 | 3d ago | |
| DUDE (test) | MultiDocFusion | ANLS17.93 | 6 | 3d ago | |
| WattBot 2025 (Private) | KohakuRAG | Score0.861 | 5 | 1mo ago | |
| WattBot 2025 (Public) | KohakuRAG | Score90.2 | 5 | 1mo ago | |
| D1 | Correct Answers20 | 2 | 1mo ago |