Metraipolis › archive › dc_dc9928ec4f77
GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks — arXiv:2610.09112
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09112
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:08:55.887Z
- SHA-256 of the archived text
5a68f76d8baa777692da95650e3cc123e2123b9746394dc6076cb34ab157f933- Stable URL of this page
https://metraipolis.com/documento/dc_dc9928ec4f77
How to cite
GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks — arXiv:2610.09112 — arXiv (en), https://arxiv.org/abs/2610.09112 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dc9928ec4f77 (Metraipolis archive dc_dc9928ec4f77, sha256 5a68f76d8baa777692da95650e3cc123e2123b9746394dc6076cb34ab157f933)
BibTeX
@misc{dcdc9928ec4f77,
title = {GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks — arXiv:2610.09112},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09112},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dc9928ec4f77; sha256 5a68f76d8baa777692da95650e3cc123e2123b9746394dc6076cb34ab157f933},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09112. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Before tool-using LLM agents are deployed in environmental and geospatial workflows, teams need evidence that an agent reliably selects the right operations against real APIs. We introduce GeoNatureAgent (GNA), a framework for pre-production evaluation of tool-using agents: a fixed sixteen-tool geospatial interface published as a Model Context Protocol (MCP) server, so the agent under test is the only variable, scored against an identical tool layer, task suite, and deterministic scorer. Its flagship instance is a 103-task benchmark (a 93-task main suite across 18 categories plus a ten-task comparison expansion) evaluated against an open, self-hostable geospatial API serving three environmental indicators across Spain and Portugal. We evaluate nine LLMs under three temperature-1.0 seeds, reporting capability and per-case cost as orthogonal axes. (1) Claude Sonnet 4 achieves the highest capability (61.7% +/- 0.7% on all 103 tasks; 60.8% on the main suite), followed closely by DeepSeek V3.2 (57.9%), while no other model exceeds 53%; (2) the cost-accuracy Pareto frontier is mostly open-weight, with DeepSeek V3.2 offering 93% of Claude's capability at 11.3x lower list-price cost; (3) under strict all-checks scoring the best model sits 24-36 points below the 85-97% reported on general-purpose GIS benchmarks, whereas per-check partial credit for the top four models (86-90%) is comparable, so much of that gap reflects scoring strictness rather than task difficulty alone. The MCP server, evaluation harness, benchmark, and API are publicly available; swapping the tool executors and task suite instantiates an equivalent benchmark for any geospatial domain. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.09112 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T10:14:05.902Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).