Benchmark Almanac

AI AgriBench

shennong-3.0-professional 98.9 tops its leaderboard GPT-5 98.1 leads the models we track
Not in model cards

Models answer 416 agronomy questions from agricultural extension publications. Three judges compare each answer with an expert-reviewed reference across four metrics.

The leaderboard has scores for 25 configurations on the full test. It lists the seven post-cutoff entries separately.

Released
January 2026
Kind
Tasks from real jobs
Level
University
Scored
Graded by a model
Answers
Open-ended
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Run by
A host
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
November 2025
Scores refreshed
November 20th, 2026
25
Entries on its own leaderboard, 13 models we track

shennong-3.0-professional tops the leaderboard with 98.9 GPT-5 leads with 98.1 among models released by November 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 98.1. This view shows only models we track, each at its best.

The newest model with a score here came out in November 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 shennong-3.0-professional 98.9
  2. 2 GPT-5 98.1
  3. 3 shennong-3.0-concise 97.6
  4. 4 cropwizard_gpt_5 97.4
  5. 5 cropwizard_gpt_5## 97.0
  6. 6 GPT-5.1 96.9
  7. 7 ext-bot-gpt-5.1 96.6
  8. 8 cropwizard_gpt_5_mini## 96.6
  9. 9 kissanai-cp-beta 96.5
  10. 10 GPT-5 mini 96.3
  11. 11 Gemini 2.5 Flash Sep 2025 96.1
  12. 12 cropwizard_gpt_5_mini 95.7
  13. 13 Claude 3.7 Sonnet 94.6
  14. 14 Gemini 3 Pro 94.4
  15. 15 Claude Opus 4.1 93.8
  16. 16 Kimi K2 Thinking 92.8
  17. 17 DeepSeek-V3 Mar 2025 91.8
  18. 18 ext-bot-gemini3 89.6
  19. 19 GPT-4o Mar 2025 87.2
  20. 20 Mistral Large 2 Nov 2024 85.6
  21. 21 gpt-oss-120b 84.4
  22. 22 Qwen2.5-72B 84.2
  23. 23 bayer-crop-science-ely 78.4
  24. 24 Tilva/PlantVillage 76.9
  25. 25 digital-green-bot 70.3

Source: Official AI AgriBench leaderboard

#EntryScoreSettingReleasedRun onLinkSource
1 shennong-3.0-professional 98.9 Feb 2026 Official leaderboard
2 GPT-5 98.1 Aug 2025 Dec 2025 Official leaderboard
3 shennong-3.0-concise 97.6 Feb 2026 Official leaderboard
4 cropwizard_gpt_5 97.4 Dec 2025 Official leaderboard
5 cropwizard_gpt_5## 97.0 Jan 2026 Official leaderboard
6 GPT-5.1 96.9 Nov 2025 Dec 2025 Official leaderboard
7 ext-bot-gpt-5.1 96.6 Dec 2025 Official leaderboard
8 cropwizard_gpt_5_mini## 96.6 Jan 2026 Official leaderboard
9 kissanai-cp-beta 96.5 Dec 2025 Official leaderboard
10 GPT-5 mini 96.3 Aug 2025 Dec 2025 Official leaderboard
11 Gemini 2.5 Flash Sep 2025 96.1 Sep 2025 Dec 2025 Official leaderboard
12 cropwizard_gpt_5_mini 95.7 Dec 2025 Official leaderboard
13 Claude 3.7 Sonnet 94.6 Feb 2025 Dec 2025 anthropic.com Official leaderboard
14 Gemini 3 Pro 94.4 Nov 2025 Dec 2025 Official leaderboard
15 Claude Opus 4.1 93.8 Aug 2025 Dec 2025 Official leaderboard
16 Kimi K2 Thinking 92.8 Nov 2025 Dec 2025 Official leaderboard
17 DeepSeek-V3 Mar 2025 91.8 Mar 2025 Dec 2025 Official leaderboard
18 ext-bot-gemini3 89.6 Dec 2025 Official leaderboard
19 GPT-4o Mar 2025 87.2 Mar 2025 Dec 2025 Official leaderboard
20 Mistral Large 2 Nov 2024 85.6 Nov 2024 Dec 2025 Official leaderboard
21 gpt-oss-120b 84.4 Aug 2025 Dec 2025 Official leaderboard
22 Qwen2.5-72B 84.2 Sep 2024 Dec 2025 Official leaderboard
23 bayer-crop-science-ely 78.4 Dec 2025 Official leaderboard
24 Tilva/PlantVillage 76.9 Jan 2026 Official leaderboard
25 digital-green-bot 70.3 Dec 2025 Official leaderboard
#ModelScoreSettingReleasedRun onLinkSource
1 GPT-5 98.1 Aug 2025 Dec 2025 Official leaderboard
2 GPT-5.1 96.9 Nov 2025 Dec 2025 Official leaderboard
3 GPT-5 mini 96.3 Aug 2025 Dec 2025 Official leaderboard
4 Gemini 2.5 Flash Sep 2025 96.1 Sep 2025 Dec 2025 Official leaderboard
5 Claude 3.7 Sonnet 94.6 Feb 2025 Dec 2025 anthropic.com Official leaderboard
6 Gemini 3 Pro 94.4 Nov 2025 Dec 2025 Official leaderboard
7 Claude Opus 4.1 93.8 Aug 2025 Dec 2025 Official leaderboard
8 Kimi K2 Thinking 92.8 Nov 2025 Dec 2025 Official leaderboard
9 DeepSeek-V3 Mar 2025 91.8 Mar 2025 Dec 2025 Official leaderboard
10 GPT-4o Mar 2025 87.2 Mar 2025 Dec 2025 Official leaderboard
11 Mistral Large 2 Nov 2024 85.6 Nov 2024 Dec 2025 Official leaderboard
12 gpt-oss-120b 84.4 Aug 2025 Dec 2025 Official leaderboard
13 Qwen2.5-72B 84.2 Sep 2024 Dec 2025 Official leaderboard

Rankings by part

We read 10 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Nutrition & fertility, Diseases and Horticulture / agronomic. GPT-5 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Nutrition & fertility

13 models we track have a score here.

  1. 1 GPT-5 98.4
  2. 2 Gemini 2.5 Flash Sep 2025 ↑2 97.0
  3. 3 GPT-5.1 ↓1 96.6
  4. 4 GPT-5 mini ↓1 96.3
  5. 5 Claude 3.7 Sonnet 95.2
  6. 6 Claude Opus 4.1 ↑1 94.8
  7. 7 Gemini 3 Pro ↓1 93.9
  8. 8 DeepSeek-V3 Mar 2025 ↑1 93.6
  9. 9 Kimi K2 Thinking ↓1 92.9
  10. 10 GPT-4o Mar 2025 88.0

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 Jan 2026
GPT-5 98.1

Source: Official AI AgriBench 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
80.0 85.0 90.0 95.0 100.0 2025 2026 80.0 85.0 90.0 95.0 100.0 2025 2026

Source: Official AI AgriBench 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

More in Economic work

All 210 in Economic work
  • JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
  • GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate. Seed2.1 Pro 87.9%
  • GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
  • AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
  • Agents' Last Exam v1.1 In use Agents complete 122 public long-horizon professional workflow tasks across 55 sub-industries. Deterministic graders check outcomes against hidden references. GPT-6.2 Astra 47.5%
  • CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Gemini 3 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (25 entries).
  • Pass Models scored 13 models we track, plus 12 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on aiagribench.org: 13 rows for models we track and 12 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