GISAgentBench - Spatial Joining and Geocoding: leaderboard
The Spatial Joining and Geocoding family of GISAgentBench (University of Florida, 2026): 96 tasks that associate records, coordinates or features across spatial and tabular datasets. Strict task success rate in percent against exact ground-truth output files under deterministic checks; higher is better.
Metric: Strict task success rate (%; share of the 96 Spatial Joining and Geocoding tasks whose output file matches the exact ground-truth output within the task's declared tolerance under deterministic structural and value checks, no LLM judge; LangChain ReAct agent over a fixed harness of 128 GIS APIs backed by QGIS, GDAL and GeoPandas; 50 tool calls per task, temperature 0, one run per task, July 2026). Source: arxiv.org. Saturation forecast: Around March 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 38.5 | #94 |
| 2 | DeepSeek V4 Pro | 31.2 | #188 |
| 3 | Claude Opus 4.6 | 30.2 | #102 |
| 4 | GPT-5.4 | 18.8 | #123 |
| 5 | Qwen 3.6 27B | 18.8 | #341 |
| 6 | GPT-OSS-120B | 12.5 | #700 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=gisagentbench-spatial-joining-and-geocoding · How It Works · Data refreshed daily, snapshot 2026-11-13.