| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MathVista GPS | DreamPRM (o4-mini) | Accuracy95.7 | 38 | 3d ago | |
| Geometry3K (test) | Human Expert | Choice Accuracy90.9 | 32 | 3d ago | |
| GeoQA | Top-1 Acc92.04 | 26 | 3d ago | ||
| Geo3K | Gemini 2.0 Flash | Top-1 Accuracy58.2 | 19 | 3d ago | |
| PGPS9K (test) | AutoGPS | Completion75.3 | 18 | 3d ago | |
| Formalgeo7k | GeoFocus-7B | Top-1 Accuracy63.5 | 17 | 3d ago | |
| GeoQA (test) | Human | Choice Accuracy92.3 | 13 | 3d ago | |
| Geometry3K 1.0 (test) | Overall Score90.9 | 12 | 3d ago | ||
| IMP-Geometry3K | Accuracy76 | 10 | 3d ago | ||
| G-MATH | BBA | Accuracy34.22 | 8 | 3d ago | |
| OlympiadBench | Accuracy75.5 | 7 | 3d ago | ||
| UniGeo Prv (test) | GOLD | Accuracy98.5 | 6 | 3d ago | |
| UniGeo CAL (test) | GOLD | Accuracy75.2 | 6 | 3d ago | |
| GEOS | Inter-GPS | Accuracy67 | 5 | 3d ago | |
| GeoCode mini (test) | Accuracy42.16 | 3 | 3d ago | ||
| Geometry3K | Accuracy79.03 | 3 | 3d ago | ||
| IMO 50 2000-2024 (test) | - | - | 0 | 3d ago |