Earth observation analysis benchmark for geospatial AI
How well AI models do earth observation analysis on satellite, aerial and map data: 2 sets with results, scored the same way for every model.
Sets
| Set | Index | Tasks of this type | Best score on the set | Models |
|---|---|---|---|---|
| Earth-Bench V2 | Spatial Agent Index | all | 80 | 48 |
| InSAR Regional Snow Mapping | Spatial Agent Index | all | 100 | 17 |
Top models
Mean of each model's best score on the 2 sets, 0 to 100, for the 17 models that have run all of them.
| Rank | Model | Developer | Score |
|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | 90 |
| 2 | Claude Opus 5 | Anthropic | 87 |
| 3 | Claude Haiku 5.5 | Anthropic | 86 |
| 4 | Claude Opus 4.8 | Anthropic | 85 |
| 5 | Claude Sonnet 5.5 | Anthropic | 85 |
| 6 | Claude Opus 4.7 | Anthropic | 85 |
| 7 | Qwen 3.8 27B | Alibaba (Qwen) | 85 |
| 8 | Claude Sonnet 5 | Anthropic | 84 |
| 9 | Step 5 Preview | StepFun | 80 |
| 10 | Muse Glimmer 30B | Meta | 78 |
Every task type · Every evaluation · Methodology
Frequently asked questions
Which AI model is best at earth observation analysis on geospatial data?
Claude Opus 5.5 leads, on the sets below. Mean of each model's best score on the 2 sets, 0 to 100, for the 17 models that have run all of them.
What is this earth observation analysis benchmark?
Every task in our sets that asks for earth observation analysis, 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.
