| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MGSM (test) | QuaSAR | Accuracy93.4 | 109 | 28d ago | |
| MT Math100 | ROSA2 | Accuracy93.6 | 64 | 3mo ago | |
| MGSM | DeepSeek V3.2 | Accuracy94.9 | 52 | 1mo ago | |
| MGSM 1.0 (test) | MindMerger-Hard | Accuracy (ru)69.6 | 35 | 3mo ago | |
| MSVAMP | Task Arithmetic | Accuracy (English)75 | 33 | 3mo ago | |
| PolyMath (test) | Qwen2.5-32B-Instruct | Accuracy (Ar)20.3 | 30 | 12d ago | |
| MGSM Thai (test) | One Step + KMM | Accuracy41.6 | 25 | 15d ago | |
| MMATH All Languages (test) | Qwen2.5-7B-Instruct + LANG | Average Score (All)28.6 | 22 | 12d ago | |
| MMATH Out-of-Domain Languages (test) | Qwen2.5-7B-Instruct + LANG | Vietnamese Accuracy30.1 | 22 | 12d ago | |
| MMATH In-Domain Languages (test) | Qwen2.5-7B-Instruct + LANG | Accuracy (Ar)26.3 | 22 | 12d ago | |
| MGSM | FLy | Speedup (de)6.08 | 20 | 3mo ago | |
| MGSM8KInstruct (test) | Accuracy (En)47.6 | 7 | 3mo ago | ||
| MCLM 1.0 (test) | MT-MATH100 Score84.89 | 7 | 3mo ago | ||
| MGSM 18 languages | SP3F-7B | Accuracy72.5 | 6 | 3mo ago |