| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| TWEEBANK V2 (test) | Stanza | F1 Score98.64 | 14 | 3d ago | |
| Wikipedia Cyrillic-rich subsets (test) | T-pro | Russian (ru)2.38 | 4 | 3d ago | |
| Code | Qwen3 | Average Tokens per Sample1,694.26 | 3 | 3d ago | |
| Spanish | Qwen3 | Average Tokens per Sample1,102.41 | 3 | 3d ago | |
| Japanese | Avg Tokens per Sample1,040.5 | 3 | 3d ago | ||
| Chinese | Qwen3 | Average Tokens per Sample914.05 | 3 | 3d ago | |
| English Reasoning | Qwen3 | Average Tokens per Sample6,192.77 | 3 | 3d ago | |
| English General | Qwen3 | Avg Tokens per Sample794.79 | 3 | 3d ago | |
| Korean Math | Qwen3 | Avg Tokens per Sample1,297.56 | 3 | 3d ago | |
| Korean Reasoning | A.X K1 | Avg Tokens per Sample5,520.06 | 3 | 3d ago | |
| Korean General | Qwen3 | Avg Tokens/Sample905.82 | 3 | 3d ago |
