Benchmark Almanac

German fruit-story antipresuppositions

GPT-5 100% leads since August 2025
Not in model cards

Models read short German stories and choose an article or quantifier that fits the number of fruits. A later task asks whether a continuation sounds appropriate.

The later study reports separate word choice and acceptability scores for 14 models using the same German items. It includes GPT-5 and Claude Sonnet 4.5.

Released
December 2023
Kind
Reasoning in many languages
Level
Everyday
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
German
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-5 #71 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
14
Entries in its paper, 9 models we track

2 models share the top score of 100% among models released by September 2025 2 models share the top score of 100% among models released by September 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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-5 100%
  2. 2 Claude Sonnet 4.5 100%
  3. 3 GPT-4.1 97%
  4. 4 Mixtral 8x7B 78%
  5. 5 OLMo 2 32B Instruct 78%
  6. 6 Phi-4-14B 75%
  7. 7 GPT-4o Mar 2025 67%
  8. 8 Prometheus-14B 64%
  9. 9 Qwen3-8B 62%
  10. 10 Mistral 7B v0.1 60%
  11. 11 Olmo-2-7B 49%
  12. 12 LLaMA-8B 45%
  13. 13 Qwen3-14B 42%
  14. 14 Olmo-2-13B 28%

Source: The German fruit-story antipresuppositions paper

#EntryScoreSettingDateSource
1 GPT-5 100% Aug 2025 Official leaderboard
2 Claude Sonnet 4.5 100% Sep 2025 Official leaderboard
3 GPT-4.1 97% Apr 2025 Official leaderboard
4 Mixtral 8x7B 78% Dec 2023 Official leaderboard
5 OLMo 2 32B Instruct 78% Mar 2025 Official leaderboard
6 Phi-4-14B 75% Official leaderboard
7 GPT-4o Mar 2025 67% Mar 2025 Official leaderboard
8 Prometheus-14B 64% Official leaderboard
9 Qwen3-8B 62% Apr 2025 Official leaderboard
10 Mistral 7B v0.1 60% Sep 2023 Official leaderboard
11 Olmo-2-7B 49% Official leaderboard
12 LLaMA-8B 45% Official leaderboard
13 Qwen3-14B 42% Apr 2025 Official leaderboard
14 Olmo-2-13B 28% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 100% Aug 2025 Official leaderboard
2 Claude Sonnet 4.5 100% Sep 2025 Official leaderboard
3 GPT-4.1 97% Apr 2025 Official leaderboard
4 Mixtral 8x7B 78% Dec 2023 Official leaderboard
5 OLMo 2 32B Instruct 78% Mar 2025 Official leaderboard
6 GPT-4o Mar 2025 67% Mar 2025 Official leaderboard
7 Qwen3-8B 62% Apr 2025 Official leaderboard
8 Mistral 7B v0.1 60% Sep 2023 Official leaderboard
9 Qwen3-14B 42% Apr 2025 Official leaderboard

4 models have held number one in the benchmark’s paper since its release in December 2023

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

Mistral AI OpenAI
Benchmark paper Passed 85% in Apr 2025, by GPT-4.1
GPT-5 100%

Source: The German fruit-story antipresuppositions paper

GPT-4.1 on the official leaderboard passed 85% in April 2025

That was 16 months after its release. By the same age, 25% of the benchmarks from 2023 had passed 85%. 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% 2024 2025 Near the ceiling (85%) GPT-5 Mixtral 8x7B Mistral 7B v0.1 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) GPT-5 Mixtral 8x7B

Source: Official German fruit-story antipresuppositions 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

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the benchmark’s paper (14 entries).
  • Pass Models scored 9 models we track, plus 5 other entries on its leaderboard.
  • Note Newest score The newest came 14 months ago, in September 2025, for Claude Sonnet 4.5. That’s expected for a saturated test.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5 (#71).
  • Note Headroom Saturated. The best is 100%, past 85% since April 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 9 rows for models we track and 5 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