| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Geometry3K (test) | BiNSGPS | Choice Accuracy95.2 | 76 | 26d ago | |
| PGPS9K (test) | BiNSGPS | Choice Accuracy92.7 | 57 | 26d ago | |
| Geometry3K | MARS-GPS | Accuracy88.8 | 41 | 3mo ago | |
| Geo3K | Draw2Think BL | Top-1 Accuracy98.2 | 41 | 2mo ago | |
| MathVista GPS | DreamPRM (o4-mini) | Accuracy95.7 | 38 | 4mo ago | |
| GeoQA | Top-1 Acc92.04 | 26 | 4mo ago | ||
| OlympiadBench | GeometryZero-7B | BoN@3 Accuracy45.69 | 17 | 3mo ago | |
| Geomverse | GeometryZero-7B | BoN@3 Accuracy18.23 | 17 | 3mo ago | |
| MathVista | GeometryZero-7B | BoN@3 Accuracy87.15 | 17 | 3mo ago | |
| Formalgeo7k | GeoFocus-7B | Top-1 Accuracy63.5 | 17 | 4mo ago | |
| GeoQA (test) | Human | Choice Accuracy92.3 | 13 | 4mo ago | |
| OlympiadBench | Accuracy75.5 | 12 | 3mo ago | ||
| Geometry3K 1.0 (test) | Overall Score90.9 | 12 | 4mo ago | ||
| UniGeo calculation (test) | Draw2Think CT | Top-1 Accuracy96.9 | 11 | 2mo ago | |
| IMP-Geometry3K | Accuracy76 | 10 | 4mo ago | ||
| G-MATH | BBA | Accuracy34.22 | 8 | 4mo ago | |
| Geometry | DAPO | Average Score50.6 | 6 | 1mo ago | |
| PGPS9K N=1000 | Draw2Think BL | Top-1 Accuracy94.5 | 6 | 2mo ago | |
| UniGeo Prv (test) | GOLD | Accuracy98.5 | 6 | 4mo ago | |
| UniGeo CAL (test) | GOLD | Accuracy75.2 | 6 | 4mo ago | |
| PyGeoX-Wild | Accuracy81 | 5 | 1mo ago | ||
| PyGeoX (Hard) | Accuracy51 | 5 | 1mo ago | ||
| PyGeoX Medium | Accuracy57 | 5 | 1mo ago | ||
| PyGeoX Easy | Qwen3-8B-RL | Accuracy62 | 5 | 1mo ago | |
| GEOS | Inter-GPS | Accuracy67 | 5 | 4mo ago |