Benchmark Almanac

AGILE Automatic Genre Identification Benchmark

GPT-5 mini 0.82 leads since August 2025
Not in model cards

Models assign one of eight genres to web texts, including news, opinion, and advertising. The test covers English and ten other languages.

The board scores 14 language model variants on both test sets. The authors share the test texts on request.

Released
June 2023
Kind
Reading in many languages
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
Public, plus a private test set
Measures
Capability
Language
Many languages
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
In views
Many languages
Strongest model tested
GPT-5 #71 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
21
Entries on its own leaderboard, 6 models we track

GPT-5 mini leads with 0.82 among models released by September 2025 GPT-5 mini leads with 0.82 among models released by September 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The source reports no overall score, so this is the plain mean of the two Macro F1 test-set scores. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in September 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-5 mini 0.82
  2. 2 GPT-5 0.82
  3. 3 X-GENRE classifier 0.80
  4. 4 Gemini 2.5 Flash Sep 2025 0.79
  5. 5 Gemini 2.5 Pro (snapshot unspecified) 0.78
  6. 6 GPT-4o (snapshot unspecified) 0.77
  7. 7 Mistral Medium 3.1 0.75
  8. 8 Llama 3.3 0.74
  9. 9 Gemma 3 0.73
  10. 10 GPT-5 nano 0.72
  11. 11 GPT-4o-mini (snapshot unspecified) 0.71
  12. 12 Qwen3 0.64
  13. 13 CORE register classifier 0.62
  14. 14 GPT-3.5-Turbo (snapshot unspecified) 0.56
  15. 15 DeepSeek-R1 0.42
  16. 16 SVC 0.37
  17. 17 GaMS-Instruct 27B 0.35
  18. 18 LOGISTICREGRESSION 0.34
  19. 19 fastText 0.31
  20. 20 COMPLEMENTNB 0.24
  21. 21 NLI zero-shot model 0.17

Source: Official AGILE Automatic Genre Identification Benchmark leaderboard

#EntryScoreSettingDateSource
1 GPT-5 mini 0.82 Aug 2025 Official leaderboard
2 GPT-5 0.82 Aug 2025 Official leaderboard
3 X-GENRE classifier 0.80 Official leaderboard
4 Gemini 2.5 Flash Sep 2025 0.79 Sep 2025 Official leaderboard
5 Gemini 2.5 Pro (snapshot unspecified) 0.78 Official leaderboard
6 GPT-4o (snapshot unspecified) 0.77 Official leaderboard
7 Mistral Medium 3.1 0.75 Aug 2025 Official leaderboard
8 Llama 3.3 0.74 Official leaderboard
9 Gemma 3 0.73 Official leaderboard
10 GPT-5 nano 0.72 Aug 2025 Official leaderboard
11 GPT-4o-mini (snapshot unspecified) 0.71 Official leaderboard
12 Qwen3 0.64 Official leaderboard
13 CORE register classifier 0.62 Official leaderboard
14 GPT-3.5-Turbo (snapshot unspecified) 0.56 Official leaderboard
15 DeepSeek-R1 0.42 Jan 2025 Official leaderboard
16 SVC 0.37 Official leaderboard
17 GaMS-Instruct 27B 0.35 Official leaderboard
18 LOGISTICREGRESSION 0.34 Official leaderboard
19 fastText 0.31 Official leaderboard
20 COMPLEMENTNB 0.24 Official leaderboard
21 NLI zero-shot model 0.17 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 mini 0.82 Aug 2025 Official leaderboard
2 GPT-5 0.82 Aug 2025 Official leaderboard
3 Gemini 2.5 Flash Sep 2025 0.79 Sep 2025 Official leaderboard
4 Mistral Medium 3.1 0.75 Aug 2025 Official leaderboard
5 GPT-5 nano 0.72 Aug 2025 Official leaderboard
6 DeepSeek-R1 0.42 Jan 2025 Official leaderboard

2 models have held number one on the official leaderboard since its release in June 2023

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

DeepSeek OpenAI
Own leaderboard Not tested since Sep 2025
GPT-5 mini 0.82

Source: Official AGILE Automatic Genre Identification Benchmark 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
0.40 0.50 0.60 0.70 0.80 0.90 2025 0.40 0.50 0.60 0.70 0.80 0.90 2025

Source: Official AGILE Automatic Genre Identification Benchmark 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 Multilingual

All 545 in Multilingual
  • MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
  • WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET. K-EXAONE 90.5%
  • MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language. Qwen3.7-Max 87%
  • CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories. Solar Open 2 90.7%
  • Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
  • MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. 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 (21 entries).
  • Pass Models scored 6 models we track, plus 15 other entries on its leaderboard.
  • Warning Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 6 rows for models we track and 15 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