Source code
github.com/The-AI-Alliance/GEO-Bench-2-Leaderb…
the official repository with harness, tasks and verifiers
Geospatial agent benchmark · not-downloaded
GEO-Bench-2 Leaderboard
Also known as: GEO-Bench-2 Leaderboard
Evaluates: model answers only
Aggregated public model-performance registry for EO foundation models.
At a glance
unknownTasks
NoAgents run real code
PartlyData + code public
YesChecked by recomputation
—Data licence
Sources
every claim on this page traces to these recordsLeaderboard
github.com/The-AI-Alliance/GEO-Bench-2-Leaderb…
the maintainers' published results, where available
- Source paper: not published, or not yet verified — we do not guess URLs
- Source data: not published, or not yet verified — we do not guess URLs
In shortAgents do not run code inside this benchmark, and answers are checked by recomputation, not by another model's opinion. Part of the evidence base (dataset or code, not both) is published. Not a data source or golden task. Depends on GEO-Bench-2 official datasets and TerraTorch or iterate workflow.
Evidence boundaries
what a score on this benchmark does and does not tell youA good score shows
Aggregated public model-performance registry for EO foundation models.
A good score does not show
Not a data source or golden task. Depends on GEO-Bench-2 official datasets and TerraTorch or iterate workflow.
Arena status
not-downloaded — Not downloaded. Comparator for baseline model selection and reporting, not a data source.
All recorded facts (22 fields)
- Evaluates
- model
- Task families
- aggregated leaderboard over submitted metrics
- Task count
- pending verification
- Difficulty
- n/a
- Geography
- n/a
- Modality
- mixed
- Input types
- submission CSV,model metadata
- Output types
- leaderboard rankings
- Execution environment
- n/a (submission-based registry)
- Ground truth method
- depends on underlying GEO-Bench-2 datasets
- Verifier method
- normalised leaderboard aggregation over submitted metrics
- Deterministic verification
- yes
- Scoring dimensions
- submitted metrics
- Aggregation formula
- normalised leaderboard aggregation
- Trials policy
- n/a
- Variance reporting
- n/a
- Data licence
- pending verification
- Code licence
- pending verification
- Contamination concerns
- depends on underlying GEO-Bench-2 datasets and TerraTorch or iterate workflow
- Reproducibility status
- medium (depends on submission integrity and upstream dataset versioning)
- First published
- pending verification
- Latest known update
- pending verification
Machine-readable record
This entry is available as JSON at /api/registry#geo-bench-2-leaderboard. See the registry endpoint.