Axis Spatial Spatial Intelligence Index

Search modelsMenu

Visual grounding benchmark for geospatial AI

How well AI models do visual grounding on satellite, aerial and map data: 1 set with results, scored the same way for every model.

Sets

Visual grounding: sets that cover it
SetIndexTasks of this typeBest score on the setModels
BigEarthNet.txtSpatial Intelligence Index606016

Top models

Best score on BigEarthNet.txt, the set most models have run, 0 to 100.

Top models on visual grounding
RankModelDeveloperScore
1Claude Opus 5.5Anthropic60
2Claude Sonnet 5.5Anthropic57
3Claude Haiku 5.5Anthropic54
4Claude Opus 5Anthropic53
5Claude Opus 4.7Anthropic49
6Step 5 PreviewStepFun49
7Claude Opus 4.8Anthropic48
8Qwen 3.8 27BAlibaba (Qwen)48
9Claude Sonnet 5Anthropic47
10Nemotron 3.5 Super VL 120BNVIDIA47

Every task type · Every evaluation · Methodology

Frequently asked questions

Which AI model is best at visual grounding on geospatial data?
Claude Opus 5.5 leads, on the sets below. Best score on BigEarthNet.txt, the set most models have run, 0 to 100.
What is this visual grounding benchmark?
Every task in our sets that asks for visual grounding, 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 Intelligence Index, as the table shows.