Spectral indices benchmark for geospatial AI
How well AI models do spectral indices 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 |
|---|---|---|---|---|
| OpenEarthAgent Sentinel-2 | Spatial Agent Index | all | 100 | 14 |
Top models
Best score on OpenEarthAgent Sentinel-2, the set most models have run, 0 to 100.
| Rank | Model | Developer | Score |
|---|---|---|---|
| 1 | Claude Haiku 4.5 | Anthropic | 100 |
| 2 | Claude Haiku 5.5 | Anthropic | 100 |
| 3 | Claude Opus 4.7 | Anthropic | 100 |
| 4 | Claude Opus 4.8 | Anthropic | 100 |
| 5 | Claude Opus 5 | Anthropic | 100 |
| 6 | Claude Opus 5.5 | Anthropic | 100 |
| 7 | Claude Sonnet 5 | Anthropic | 100 |
| 8 | Claude Sonnet 5.5 | Anthropic | 100 |
| 9 | Nemotron 3 Ultra | NVIDIA | 100 |
| 10 | Step 5 Preview | StepFun | 100 |
Every task type · Every evaluation · Methodology
Frequently asked questions
Which AI model is best at spectral indices on geospatial data?
Claude Haiku 4.5 leads, on the sets below. Best score on OpenEarthAgent Sentinel-2, the set most models have run, 0 to 100.
What is this spectral indices benchmark?
Every task in our sets that asks for spectral indices, 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.
