University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

claude-fable-5-1

Anthropic
average±1.7 (95%)
89.0
leaderboard rankof 114
≈1
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$128
seconds per inputmean response time
23.0
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    82.0

    Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.

  • Document understanding

    96.2

    Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.

  • Information extraction

    88.5

    Best of 107 models with all its benchmarks: grok-4.6, 90.8.

  • Transcription

    94.8

    Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.

By benchmark

BenchmarkScoreThe best current run on each benchmark, on the benchmark's metric scaled to 0–100. ± is the 95% sampling margin, from resampling the run's inputs.RankRank among all current models on that benchmark; ≈ for models that cannot be told apart within sampling error.Per 1,000 docss / inputBest run
Bibliographic Data 99.3±0.4 ≈1of 115 $184$0.92 run 27.95 T1675 · 29 Sep 2026prompt.txt
Blacklist Cards 96.7±2.2 ≈8of 114 $43$1.41 run 7.66 T1682 · 17 Sep 2026prompt.txt
Book Advert XML Repair 66.6±9.8 ≈1of 119 $125$6.27 run 30.51 T1685 · 17 Sep 2026prompt.txt
Business Letters 61.0±7.2 ≈24of 114 $80$4.58 run 8.24 T1676 · 17 Sep 2026prompt.txt
Company Lists 85.1±13.1 ≈29of 114 $124$1.86 run 17.82 T1683 · 29 Sep 2026prompt.txt
Duty Rosters 94.5±2.0 ≈5of 107 $187$3.92 run 40.59 T1689 · 29 Sep 2026prompt.txt
Fraktur Adverts 99.5±0.1 ≈1of 114 $293$1.46 run 64.44 T1679 · 17 Sep 2026prompt_optimized.txt
General Meeting Minutes 94.0±3.8 ≈1of 108 $167$1.51 run 27.77 T1688 · 29 Sep 2026prompt.txt
Library Cards 89.3±1.3 ≈15of 114 $48$13 run 7.77 T1680 · 29 Sep 2026prompt.txt
Magazine Pages 96.8±2.8 ≈1of 115 $66$3.04 run 8.94 T1687 · 29 Sep 2026prompt.txt
Medieval Manuscripts 87.6±9.0 ≈1of 114 $95$1.14 run 19.9 T1681 · 17 Sep 2026prompt.txt
Personnel Cards 97.5±0.7 ≈1of 114 $128$7.82 run 14.4 T1686 · 29 Sep 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
8 Sep 202610.050.0archived price page

Every run

  1. T1689Duty Rosters29 Sep 202694.5$3.92
  2. T1688General Meeting Minutes29 Sep 202694.0$1.51
  3. T1687Magazine Pages29 Sep 202696.8$3.04
  4. T1686Personnel Cards29 Sep 202697.5$7.82
  5. T1685Book Advert XML Repair29 Sep 202660.8$6.91
  6. T1684Company Lists29 Sep 202669.1$2.10
  7. T1683Company Lists29 Sep 202685.1$1.86
  8. T1682Blacklist Cards29 Sep 202696.4$1.84
  9. T1681Medieval Manuscripts29 Sep 202686.9$1.41
  10. T1680Library Cards29 Sep 202689.3$13
  11. T1679Fraktur Adverts29 Sep 202699.5$1.52
  12. T1678Business Letters29 Sep 2026 · 32 signed letters65.0$3.55
  13. T1677Business Letters29 Sep 2026 · 25 unsigned letters61.0$2.13
  14. T1676Business Letters29 Sep 2026 · all 57 letters60.0$5.62
  15. T1675Bibliographic Data29 Sep 202699.3$0.92
  16. T1689Duty Rosters17 Sep 20260.3$3.43
  17. T1688General Meeting Minutes17 Sep 202619.7$1.47
  18. T1687Magazine Pages17 Sep 20260.0$3.08
  19. T1686Personnel Cards17 Sep 202697.5$7.70
  20. T1685Book Advert XML Repair17 Sep 202666.6$6.27
  21. T1684Company Lists17 Sep 20260.0$1.66
  22. T1683Company Lists17 Sep 202684.8$1.76
  23. T1682Blacklist Cards17 Sep 202696.7$1.41
  24. T1681Medieval Manuscripts17 Sep 202687.6$1.14
  25. T1680Library Cards17 Sep 202689.0$6.96
  26. T1679Fraktur Adverts17 Sep 202699.5$1.46
  27. T1678Business Letters17 Sep 2026 · 32 signed letters65.0$2.90
  28. T1677Business Letters17 Sep 2026 · 25 unsigned letters58.0$1.61
  29. T1676Business Letters17 Sep 2026 · all 57 letters61.0$4.58
  30. T1675Bibliographic Data17 Sep 202646.0$0.93