GISAgentBench - Spatial Pattern Analysis: leaderboard

Metric: Strict task success rate (%; share of the 52 Spatial Pattern Analysis 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 2029. 6 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)23.1
2Claude Opus 4.623.1
3DeepSeek V4 Pro21.2
4GPT-5.417.3
5Qwen 3.6 27B13.5
6GPT-OSS-120B11.5

Interactive version: theaggregate.ai/benchmark?slug=gisagentbench-spatial-pattern-analysis · How It Works · Data refreshed daily, snapshot 2026-09-29.