Benchmark Almanac

BIG-bench Understanding Fables

PaLM 535b 77.3% tops its leaderboard Gopher (280B) 39.6% leads the models we track
Not in model cards

Models read a paraphrased Aesop fable and choose its moral from five options. The test contains 189 fables.

Google reports five-shot scores of 75.7 for PaLM and 78.4 for U-PaLM. The author also releases the task for zero-shot testing.

Made by
Denis Emelin
Released
June 2021
Kind
Reading comprehension
Level
Everyday
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Last model added
December 2021
Scores refreshed
November 20th, 2026
51
Entries on its own leaderboard, 2 models we track

PaLM 535b tops the leaderboard with 77.3% Gopher (280B) leads with 39.6%

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 Gopher (280B), at 39.6%. This view shows only models we track, each at its best.

  1. 1 PaLM 535b 2-shot 77.3%
  2. 2 Gopher (280B) 5-shot 39.6%
  3. 3 PaLM 64b 5-shot 32.3%
  4. 4 PaLM 8b 5-shot 29.6%
  5. 5 BIG-G T=0 2m 0-shot, temperature 0 27%
  6. 6 BIG-G T=1 2m 0-shot, temperature 1 27%
  7. 7 GPT-3 200B 3-shot, temperature 0 27%
  8. 8 BIG-G T=0 53m 2-shot, temperature 0 26.5%
  9. 9 BIG-G T=1 53m 2-shot, temperature 1 26.5%
  10. 10 GPT-3 13B 3-shot, temperature 0 26.5%
  11. 11 BIG-G T=0 8b 3-shot, temperature 0 25.9%
  12. 12 BIG-G T=1 8b 3-shot, temperature 1 25.9%
  13. 13 BIG-G T=0 1b 2-shot, temperature 0 25.4%
  14. 14 BIG-G T=1 1b 2-shot, temperature 1 25.4%
  15. 15 BIG-G T=0 128b 1-shot, temperature 0 24.9%
  16. 16 BIG-G T=1 128b 1-shot, temperature 1 24.9%
  17. 17 GPT-3 Small 1-shot, temperature 0 24.3%
  18. 18 BIG-G sparse 1b 3-shot, temperature 0 23.8%
  19. 19 BIG-G sparse 244m 2-shot, temperature 0 23.8%
  20. 20 BIG-G sparse 2b 1-shot, temperature 0 23.8%
  21. 21 BIG-G sparse 2m 3-shot, temperature 0 23.8%
  22. 22 BIG-G sparse 422m 3-shot, temperature 0 23.8%
  23. 23 BIG-G sparse 53m 0-shot, temperature 0 23.8%
  24. 24 BIG-G T=0 422m 3-shot, temperature 0 23.8%
  25. 25 BIG-G T=1 422m 3-shot, temperature 1 23.8%
  26. 26 BIG-G sparse 16m 0-shot, temperature 0 22.8%
  27. 27 BIG-G sparse 8b 3-shot, temperature 0 22.8%
  28. 28 BIG-G T=0 244m 2-shot, temperature 0 22.8%
  29. 29 BIG-G T=0 2b 0-shot, temperature 0 22.8%
  30. 30 BIG-G T=1 244m 2-shot, temperature 1 22.8%
  31. 31 BIG-G T=1 2b 0-shot, temperature 1 22.8%
  32. 32 Gopher 417M 5-shot 22.5%
  33. 33 BIG-G T=0 27b 3-shot, temperature 0 22.2%
  34. 34 BIG-G T=1 27b 3-shot, temperature 1 22.2%
  35. 35 BIG-G sparse 4b 3-shot, temperature 0 21.7%
  36. 36 BIG-G T=0 16m 1-shot, temperature 0 21.7%
  37. 37 BIG-G T=1 16m 1-shot, temperature 1 21.7%
  38. 38 BIG-G sparse 125m 0-shot, temperature 0 21.2%
  39. 39 BIG-G T=0 4b 3-shot, temperature 0 21.2%
  40. 40 BIG-G T=1 4b 3-shot, temperature 1 21.2%
  41. 41 Gopher 44M 5-shot 20.3%
  42. 42 GPT-3 3B 2-shot, temperature 0 20.1%
  43. 43 GPT-3 Medium 1-shot, temperature 0 20.1%
  44. 44 Gopher 7.1B 5-shot 19.8%
  45. 45 BIG-G T=0 125m 3-shot, temperature 0 19.6%
  46. 46 BIG-G T=1 125m 3-shot, temperature 1 19.6%
  47. 47 GPT-3 6B 2-shot, temperature 0 19.6%
  48. 48 GPT-3 XL 1-shot, temperature 0 19.6%
  49. 49 Gopher 117M 5-shot 18.7%
  50. 50 GPT-3 Large 0-shot, temperature 0 18%
  51. 51 Gopher 1.4B 5-shot 17.1%

Source: Official BIG-bench Understanding Fables leaderboard

Source: Official BIG-bench Understanding Fables leaderboard

#EntryScoreSettingDateLinkSource
1 PaLM 535b 535B 77.3% 2-shot raw.githubusercontent.com Official leaderboard
2 Gopher (280B) 39.6% 5-shot Dec 2021 raw.githubusercontent.com Official leaderboard
3 PaLM 64b 62.5B 32.3% 5-shot raw.githubusercontent.com Official leaderboard
4 PaLM 8b 8.6B 29.6% 5-shot raw.githubusercontent.com Official leaderboard
5 BIG-G T=0 2m 0B 27% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
6 BIG-G T=1 2m 0B 27% 0-shot, temperature 1 raw.githubusercontent.com Official leaderboard
7 GPT-3 200B 200B 27% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
8 BIG-G T=0 53m 0.1B 26.5% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
9 BIG-G T=1 53m 0.1B 26.5% 2-shot, temperature 1 raw.githubusercontent.com Official leaderboard
10 GPT-3 13B 13B 26.5% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
11 BIG-G T=0 8b 8.7B 25.9% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
12 BIG-G T=1 8b 8.7B 25.9% 3-shot, temperature 1 raw.githubusercontent.com Official leaderboard
13 BIG-G T=0 1b 1.1B 25.4% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
14 BIG-G T=1 1b 1.1B 25.4% 2-shot, temperature 1 raw.githubusercontent.com Official leaderboard
15 BIG-G T=0 128b 137.7B 24.9% 1-shot, temperature 0 raw.githubusercontent.com Official leaderboard
16 BIG-G T=1 128b 137.7B 24.9% 1-shot, temperature 1 raw.githubusercontent.com Official leaderboard
17 GPT-3 Small 0.1B 24.3% 1-shot, temperature 0 raw.githubusercontent.com Official leaderboard
18 BIG-G sparse 1b 7.6B 23.8% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
19 BIG-G sparse 244m 2.8B 23.8% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
20 BIG-G sparse 2b 17.3B 23.8% 1-shot, temperature 0 raw.githubusercontent.com Official leaderboard
21 BIG-G sparse 2m 0.1B 23.8% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
22 BIG-G sparse 422m 4.1B 23.8% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
23 BIG-G sparse 53m 0.5B 23.8% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
24 BIG-G T=0 422m 0.5B 23.8% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
25 BIG-G T=1 422m 0.5B 23.8% 3-shot, temperature 1 raw.githubusercontent.com Official leaderboard
26 BIG-G sparse 16m 0.2B 22.8% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
27 BIG-G sparse 8b 60.3B 22.8% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
28 BIG-G T=0 244m 0.3B 22.8% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
29 BIG-G T=0 2b 2.2B 22.8% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
30 BIG-G T=1 244m 0.3B 22.8% 2-shot, temperature 1 raw.githubusercontent.com Official leaderboard
31 BIG-G T=1 2b 2.2B 22.8% 0-shot, temperature 1 raw.githubusercontent.com Official leaderboard
32 Gopher 417M 0.5B 22.5% 5-shot raw.githubusercontent.com Official leaderboard
33 BIG-G T=0 27b 29.2B 22.2% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
34 BIG-G T=1 27b 29.2B 22.2% 3-shot, temperature 1 raw.githubusercontent.com Official leaderboard
35 BIG-G sparse 4b 25.5B 21.7% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
36 BIG-G T=0 16m 0B 21.7% 1-shot, temperature 0 raw.githubusercontent.com Official leaderboard
37 BIG-G T=1 16m 0B 21.7% 1-shot, temperature 1 raw.githubusercontent.com Official leaderboard
38 BIG-G sparse 125m 1.8B 21.2% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
39 BIG-G T=0 4b 3.7B 21.2% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
40 BIG-G T=1 4b 3.7B 21.2% 3-shot, temperature 1 raw.githubusercontent.com Official leaderboard
41 Gopher 44M 0.1B 20.3% 5-shot raw.githubusercontent.com Official leaderboard
42 GPT-3 3B 3B 20.1% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
43 GPT-3 Medium 20.1% 1-shot, temperature 0 Jun 2020 raw.githubusercontent.com Official leaderboard
44 Gopher 7.1B 7.2B 19.8% 5-shot raw.githubusercontent.com Official leaderboard
45 BIG-G T=0 125m 0.2B 19.6% 3-shot, temperature 0 raw.githubusercontent.com Official leaderboard
46 BIG-G T=1 125m 0.2B 19.6% 3-shot, temperature 1 raw.githubusercontent.com Official leaderboard
47 GPT-3 6B 6B 19.6% 2-shot, temperature 0 raw.githubusercontent.com Official leaderboard
48 GPT-3 XL 1.3B 19.6% 1-shot, temperature 0 raw.githubusercontent.com Official leaderboard
49 Gopher 117M 0.1B 18.7% 5-shot raw.githubusercontent.com Official leaderboard
50 GPT-3 Large 0.8B 18% 0-shot, temperature 0 raw.githubusercontent.com Official leaderboard
51 Gopher 1.4B 1.5B 17.1% 5-shot raw.githubusercontent.com Official leaderboard
#ModelScoreSettingDateLinkSource
1 Gopher (280B) 39.6% 5-shot Dec 2021 raw.githubusercontent.com Official leaderboard
2 GPT-3 Medium 20.1% 1-shot, temperature 0 Jun 2020 raw.githubusercontent.com Official leaderboard

Number one over time

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

Google DeepMind
Own leaderboard Not tested since Dec 2021

Source: Official BIG-bench Understanding Fables 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% 2020 2021 2022 Near the ceiling (85%) Gopher (280B) GPT-3 Medium 0% 20% 40% 60% 80% 100% 2020 2021 2022 Near the ceiling (85%) Gopher (280B) GPT-3 Medium

Source: Official BIG-bench Understanding Fables 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 4 years ago, in December 2021, for Gopher (280B). 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 (51 entries).
  • Pass Models scored 2 models we track, plus 49 other entries on its leaderboard.
  • Warning Newest score The newest came 4 years ago, in December 2021, for Gopher (280B).
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 39.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 5 rows for models we track and 49 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