| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MMMLU | Accuracy87.2 | 18 | 4mo ago | ||
| MMLU ProX | Qwen-3.5 | Accuracy86.4 | 10 | 1mo ago | |
| CMMLU | KEEL | CMMLU Score71.8 | 6 | 3mo ago | |
| INCLUDE | DeepSeek-V3.1 Base | Accuracy77.2 | 4 | 4mo ago | |
| GlobalMMLU | DeepSeek-V3.2 Exp Base | Accuracy82 | 4 | 4mo ago |