How well AI understands the physical world, at planetary scale
We run the Spatial Intelligence Index: AI models benchmarked on real geospatial work, with every score and cost published. Then we build spatial workflows and agents with them.
Try:
What we do
See the Spatial Intelligence Index
40 AI models benchmarked on the same geospatial tasks, with the score, cost per task and failures for each. It is free to read.
See the indexProduction geospatial workflows
We take a GIS or Earth observation task your team does by hand, such as mapping, change detection or checking map data against imagery, and build it into a pipeline that runs on its own and checks its output.
Tell us about the workSpatial agents
We build agents that read maps, imagery and GIS files and finish the work: find an object in street or satellite images, check a map against what the camera sees, answer questions over your data. We choose the model with the index.
Tell us what it should doEvaluate models on your own tasks
We run the models on tasks from your work, with your data, under the same checks as the Spatial Intelligence Index. You get the score, cost per task and failures for each.
Tell us about your tasksFor AI labs: evaluation tasks, training data and environments
We write geospatial evaluation tasks, training data and environments, each with a verifier that checks the result.
See what we offer AI labsWork we have done
Grounding models that find traffic signs in street images, LLM-drafted labels for tents and buildings in satellite images, a tool that finds the coordinate system of a GIS file, and the index itself. Each one names the model, the data and the result.
See our workBring us one piece of spatial work.
Tell us one GIS or imagery task your team does today and the data it uses. We tell you which model can do it, what it costs per task, and build it if you want.
Work with us

























