Benchmark Almanac

GeoBenchX

GPT-4o Aug 2024 51.5% leads since March 2025
Not in model cards
Version 1

Agents use GIS tools to solve spatial analysis tasks or identify missing data. Three model judges compare the proposed workflows with reference solutions.

The first release tested seven models on 202 tasks. The authors classified 104 tasks as solvable and 98 as unsolvable.

Released
March 2025
Kind
Apps and MCP servers
Level
University
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Maps and geodata → Actions, Text
Interaction
Agentic
Environment
APIs and tools
Items from
Generated
Run by
Anyone
Occupations
Architecture and engineering
In views
Agents
Strongest model tested
Claude 3.7 Sonnet #119 on the capability index
Last model added
February 2025
Scores refreshed
November 20th, 2026
7
Models scored in its paper

GPT-4o leads with 51.5% among models released by February 2025

Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the plain mean of full-match rates on solvable and unsolvable tasks. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in February 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-4o Aug 2024 51.5%
  2. 2 o3-mini 48.5%
  3. 3 Claude 3.5 Sonnet (October 2024) 45%
  4. 4 Gemini 2.0 Flash Feb 2025 44.5%
  5. 5 GPT-4o mini 39%
  6. 6 Claude 3.7 Sonnet 32%
  7. 7 Claude 3.5 Haiku 31.5%

Source: The GeoBenchX paper

#ModelScoreSettingDateSource
1 GPT-4o Aug 2024 51.5% Aug 2024 Official leaderboard
2 o3-mini 48.5% Jan 2025 Official leaderboard
3 Claude 3.5 Sonnet (October 2024) 45% Oct 2024 Official leaderboard
4 Gemini 2.0 Flash Feb 2025 44.5% Feb 2025 Official leaderboard
5 GPT-4o mini 39% Jul 2024 Official leaderboard
6 Claude 3.7 Sonnet 32% Feb 2025 Official leaderboard
7 Claude 3.5 Haiku 31.5% Oct 2024 Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s paper, each ranking models on one part of the test, such as Unsolvable tasks, Solvable tasks and Merge-visualize. o3-mini leads 2 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Unsolvable tasks

7 models we track have a score here.

  1. 1 o3-mini ↑1 76%
  2. 2 GPT-4o mini ↑3 62%
  3. 3 GPT-4o Aug 2024 ↓2 59%
  4. 4 Gemini 2.0 Flash Feb 2025 46%
  5. 5 Claude 3.5 Sonnet (October 2024) ↓2 37%
  6. 6 Claude 3.5 Haiku ↑1 20%
  7. 7 Claude 3.7 Sonnet ↓1 14%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI
Benchmark paper Replaced by v2, Jun 2025
GPT-4o 51.5%

Source: The GeoBenchX paper

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4o GPT-4o mini 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4o GPT-4o mini

Source: Official GeoBenchX leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

OpenAI Meta Anthropic Google DeepMind
GeoBenchX v2 8 leaders
GeoBenchX Replaced by v2, Jun 2025
GPT-4o 51.5%
GeoBenchX v2 7 leaders

Source: Official leaderboards

More in Tool use

All 265 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Neon 56.3%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (7 entries).
  • Pass Models scored 7 models we track.
  • Note Newest score The newest came 21 months ago, in February 2025, for Claude 3.7 Sonnet. That’s expected for a test replaced by GeoBenchX v2.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude 3.7 Sonnet (#119).
  • Pass Headroom The best is 51.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 7 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check