Benchmark Almanac

The Aggregate Long-tail Book Recall

Claude Fable 5.2 77% leads since November 2026
Not in model cards

Models answer closed-book questions about 60 Project Gutenberg books. The scorer checks the answer text, and a second reviewer can accept equivalent wording.

The maker ranks 21 models on 282 questions. The books span five levels of Wikipedia readership. Some models answered fewer than 282 questions.

Kind
Trivia and open questions
Level
School
Scored
Checked automatically
Answers
Short answer
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
21
Models scored on its own leaderboard

Claude Fable 5.2 leads with 77%

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

  1. 1 Claude Fable 5.2 high 77%
  2. 2 Claude Fable 5.1 high 69.5%
  3. 3 Claude Fable 5 high 69.1%
  4. 4 Claude Opus 5 high 62.8%
  5. 5 Gemini 3.7 Flash high 56.7%
  6. 6 GPT-6.2 Astra high 56.1%
  7. 7 GPT-5.6 Sol high 55.3%
  8. 8 GPT-6 Astra high 50.7%
  9. 9 GPT-6.2 Sol high 48.8%
  10. 10 GPT-5.4 high 46.1%
  11. 11 DeepSeek-V4-Pro high 44%
  12. 12 Claude Opus 4.8 high 42.2%
  13. 13 Muse Spark 1.3 high 39.4%
  14. 14 GPT-5.4 Mini high 38.3%
  15. 15 GPT-5.2 high 37.2%
  16. 16 Qwen 3.6 Plus high 31.9%
  17. 17 Kimi K2.6 high 30.5%
  18. 18 Kimi K2 Sep 2025, high 28.7%
  19. 19 Qwen 3.6 35B-A3B high 27.7%
  20. 20 gpt-oss-120b high 17.4%
  21. 21 Claude Haiku 4.5 high 17.4%

Source: Official The Aggregate Long-tail Book Recall leaderboard

#ModelScoreSettingDateSource
1 Claude Fable 5.2 77% high Nov 2026 Official leaderboard
2 Claude Fable 5.1 69.5% high Sep 2026 Official leaderboard
3 Claude Fable 5 69.1% high Jun 2026 Official leaderboard
4 Claude Opus 5 62.8% high Jul 2026 Official leaderboard
5 Gemini 3.7 Flash 56.7% high Aug 2026 Official leaderboard
6 GPT-6.2 Astra 56.1% high Nov 2026 Official leaderboard
7 GPT-5.6 Sol 55.3% high Jul 2026 Official leaderboard
8 GPT-6 Astra 50.7% high Sep 2026 Official leaderboard
9 GPT-6.2 Sol 48.8% high Nov 2026 Official leaderboard
10 GPT-5.4 46.1% high Mar 2026 Official leaderboard
11 DeepSeek-V4-Pro 44% high Apr 2026 Official leaderboard
12 Claude Opus 4.8 42.2% high May 2026 Official leaderboard
13 Muse Spark 1.3 39.4% high Sep 2026 Official leaderboard
14 GPT-5.4 Mini 38.3% high Mar 2026 Official leaderboard
15 GPT-5.2 37.2% high Dec 2025 Official leaderboard
16 Qwen 3.6 Plus 31.9% high Mar 2026 Official leaderboard
17 Kimi K2.6 30.5% high Apr 2026 Official leaderboard
18 Kimi K2 Sep 2025 28.7% high Sep 2025 Official leaderboard
19 Qwen 3.6 35B-A3B 27.7% high Apr 2026 Official leaderboard
20 gpt-oss-120b 17.4% high Aug 2025 Official leaderboard
21 Claude Haiku 4.5 17.4% high Oct 2025 Official leaderboard

7 models have held number one on the official leaderboard since 2023

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

OpenAI Moonshot AI Anthropic
Own leaderboard 7 leaders

Source: Official The Aggregate Long-tail Book Recall 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% 2026 2027 Near the ceiling (85%) Claude Fable 5.2 Claude Fable 5.1 GPT-5.4 GPT-5.2 Kimi K2 gpt-oss-120b 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Fable 5.2 Claude Fable 5.1

Source: Official The Aggregate Long-tail Book Recall 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 Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

Data audit

10 of 11 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (21 entries).
  • Pass Models scored 21 models we track.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Pass Headroom The best is 77%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on theaggregate.ai: 21 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