Spatial reasoning benchmark for geospatial AI
How well AI models do spatial reasoning on satellite, aerial and map data: 1 set with results, scored the same way for every model.
Sets
| Set | Index | Tasks of this type | Best score on the set | Models |
|---|---|---|---|---|
| MultiGlobeQA | Spatial Agent Index | all | 80 | 30 |
Top models
Best score on MultiGlobeQA, the set most models have run, 0 to 100.
| Rank | Model | Developer | Score |
|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | 80 |
| 2 | Claude Sonnet 5.5 | Anthropic | 80 |
| 3 | Claude Opus 5 | Anthropic | 79 |
| 4 | Claude Fable 5.1 | Anthropic | 79 |
| 5 | DeepSeek V4.1 Flash | DeepSeek | 78 |
| 6 | Claude Haiku 5.5 | Anthropic | 77 |
| 7 | Kimi K3 | Moonshot AI | 77 |
| 8 | Claude Opus 4.7 | Anthropic | 77 |
| 9 | GLM-5.3-Flash | Z.ai | 76 |
| 10 | Claude Opus 4.8 | Anthropic | 75 |
Every task type · Every evaluation · Methodology
Frequently asked questions
Which AI model is best at spatial reasoning on geospatial data?
Claude Opus 5.5 leads, on the sets below. Best score on MultiGlobeQA, the set most models have run, 0 to 100.
What is this spatial reasoning benchmark?
Every task in our sets that asks for spatial reasoning, in GIS and Earth observation work, with each set's own score. It is a view over the sets, not an index of its own.
Which index does it count in?
Each set counts in one index: the Spatial Agent Index, as the table shows.
