| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| GeoBrowse Level 2 BrowseComp-style | GATE | City Accuracy26.3 | 24 | 3mo ago | |
| GeoBrowse Level 1 Visual Cues | GATE | City Accuracy24 | 24 | 3mo ago | |
| WanderBench Country level | Gemini-2.5 Pro | Accuracy95.8 | 21 | 4mo ago | |
| WanderBench City level | Gemini-2.5 Pro | Accuracy78.3 | 21 | 4mo ago | |
| WanderBench Street level | Accuracy52.4 | 21 | 4mo ago | ||
| WanderBench Overall | Gemini-2.5 Pro | Distance Error (km)134.3 | 21 | 4mo ago | |
| WABAD (test) | Annotated Species | Median Error (km)736 | 11 | 1mo ago | |
| AVG (test) | Ours | City Acc (25km)8.3 | 10 | 4mo ago | |
| GeoSeek (val) | GeoAgent | Success Rate (City 25km)15.69 | 9 | 5mo ago | |
| DOXBENCH Difficult 1.0 | GEO-Detective | Country Accuracy100 | 8 | 5mo ago | |
| DOXBENCH Moderate 1.0 | Country Acc99.2 | 8 | 5mo ago | ||
| DOXBENCH Easy 1.0 | Country Accuracy100 | 8 | 5mo ago | ||
| MP16-pro Extremely Difficult | GEO-Detective | Unknown Rate50 | 2 | 5mo ago | |
| MP16-pro Very Difficult | GEO-Detective | Unknown Rate28.9 | 2 | 5mo ago | |
| MP16-pro Difficult | GEO-Detective | Unknown Rate22.6 | 2 | 5mo ago | |
| MP16-pro Moderate | GEO-Detective | Unknown Rate0.263 | 2 | 5mo ago | |
| MP16-pro Easy | GEO-Detective | Unknown Rate0.186 | 2 | 5mo ago |