Benchmark Almanac

OpenEarthAgent

OpenEarthAgent 4B 45.6% tops its leaderboard GPT-5 44.5% leads the models we track
Not in model cards

Agents use tools such as QGIS to measure features in satellite images and maps. The authors grade tool decisions and final answers.

Axis Spatial scores 17 distinct models on an image subset. The authors publish 1,169 evaluation cases and separate step prediction from live tool use.

Released
February 2026
Kind
Apps and MCP servers
Level
University
Scored
Graded by a model
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
Interaction
Agentic
Environment
QGIS
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
In views
Agents, Vision
Strongest model tested
GPT-5 #71 on the capability index
Last model added
August 2025
Scores refreshed
November 20th, 2026
10
Entries on its own leaderboard, 3 models we track

OpenEarthAgent 4B tops the leaderboard with 45.6% GPT-5 leads with 44.5% among models released by August 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is GPT-5, at 44.5%. This view shows only models we track, each at its best.

The newest model with a score here came out in August 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 OpenEarthAgent 4B 45.6%
  2. 2 GPT-5 44.5%
  3. 3 GPT-4o Mar 2025 39.7%
  4. 4 o4-mini 35.9%
  5. 5 Qwen2.5-Instruct 7B 16.6%
  6. 6 Qwen3-Instruct-2507 4B 13.7%
  7. 7 Llama-3.1-Instruct 8B 12.9%
  8. 8 Qwen2.5-Instruct 3B 1.60%
  9. 9 Mistral-Instruct-v0.3 7B 0.48%
  10. 10 InternLM3-Instruct 8B 0.24%

Source: Official OpenEarthAgent leaderboard

Source: Official OpenEarthAgent leaderboard

#EntryScoreSettingDateSource
1 OpenEarthAgent 4B 4B 45.6% Official leaderboard
2 GPT-5 44.5% Aug 2025 Official leaderboard
3 GPT-4o Mar 2025 39.7% Mar 2025 Official leaderboard
4 o4-mini 35.9% Apr 2025 Official leaderboard
5 Qwen2.5-Instruct 7B 7B 16.6% Official leaderboard
6 Qwen3-Instruct-2507 4B 4B 13.7% Official leaderboard
7 Llama-3.1-Instruct 8B 8B 12.9% Official leaderboard
8 Qwen2.5-Instruct 3B 3B 1.60% Official leaderboard
9 Mistral-Instruct-v0.3 7B 7B 0.48% Official leaderboard
10 InternLM3-Instruct 8B 8B 0.24% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 44.5% Aug 2025 Official leaderboard
2 GPT-4o Mar 2025 39.7% Mar 2025 Official leaderboard
3 o4-mini 35.9% Apr 2025 Official leaderboard

Number one over time

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

OpenAI
Own leaderboard Not tested since Feb 2026
GPT-5 44.5%

Source: Official OpenEarthAgent leaderboard

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
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-5 GPT-4o 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-5

Source: Official OpenEarthAgent 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.

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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 15 months ago, in August 2025, for GPT-5. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (10 entries).
  • Pass Models scored 3 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 15 months ago, in August 2025, for GPT-5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Pass Headroom The best is 44.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 3 rows for models we track and 7 other entries 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