[Research Article] GeoJSON agents: A multi-agent LLM architecture for geospatial analysis—function calling vs. code generation

A new study published in Big Earth Data proposes GeoJSON agents, a novel multi-agent large language model (LLM) framework for geospatial analysis that transforms natural language instructions into structured GeoJSON operations through function calling and code generation. Experiments on a hierarchical benchmark of 70 spatial tasks show that the code generation–based agent achieved 97.14% accuracy and the function calling–based agent achieved 85.71%, both significantly outperforming general-purpose models, while highlighting the trade-off between flexibility and execution stability in GeoAI applications. The data that support the findings of this study are openly available in figshare at https://doi.org/10.6084/m9.figshare.29921492.

Citation

Luo, Q., Lin, Q., Xu, L., Wu, S., Mao, R., Wang, C., … Du, Z. (2026). GeoJSON agents: a multi-agent LLM architecture for geospatial analysis—function calling vs. code generation. Big Earth Data, 1–55. https://doi.org/10.1080/20964471.2026.2615511

Abstract

Large Language Models (LLMs) have demonstrated substantial progress in task automation and natural language understanding. However, without domain expertise in geographic information science (GIS), they continue to encounter limitations including reduced accuracy and unstable performance when processing complex spatial tasks. To address these challenges, we propose GeoJSON agents—a novel multi-agent LLM architecture specifically designed for geospatial analysis. This framework transforms natural language instructions into structured GeoJSON operations through two widely adopted LLM enhancement techniques: function calling and code generation. The architecture integrates three core components: task parsing, agent collaboration, and result integration. The planner agent systematically decomposes user-defined tasks into executable subtasks, while specialized worker agents perform spatial data processing and analysis either by invoking predefined function APIs or by dynamically generating and executing Python-based analytical code. The system produces reusable, standards-compliant GeoJSON outputs through iterative refinement. To systematically evaluate both approaches, we constructed a hierarchical benchmark comprising 70 tasks spanning basic, intermediate, and advanced complexity levels, conducting experiments with OpenAI’s GPT-4o as the core model. Results indicate that the code generation–based agent achieved 97.14% accuracy, while the function calling–based agent attained 85.71%—both significantly outperforming the best-performing general-purpose model (48.57%). Comparative analysis reveals that code generation offers superior flexibility for complex, open-ended tasks, whereas function calling provides enhanced execution stability for structured operations. This study represents the first systematic integration of GeoJSON data with a multi-agent LLM framework and provides empirical evidence comparing two mainstream enhancement methodologies in geospatial contexts, offering new perspectives for improving GeoAI system performance and reducing barriers to GIS application.

#geoscience #remote sensing #earth observation #GIS #data analysis #Big Data #visualization

Big Earth Data is an interdisciplinary Open Access journal which aims to provide an efficient and high-quality platform for promoting the sharing, processing and analyses of Earth-related big data, thereby revolutionizing the cognition of the Earth’s systems. The journal publishes a wide range of content, including Research Articles, Review Articles, Data Notes, Technical Notes, and Perspectives. It is now included in ESCI (IF=3.8, Q1), Scopus (CiteScore=9.0, Q1), Ei Compendex, GEOBASE, and Inspec. Starting from 2023, Big Earth Data has announced a new award series for authors: Best and Outstanding Paper Awards.