University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

claude-opus-5

Anthropic
average±3.4 (95%)
72.5
leaderboard rankof 114
≈54
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$48
seconds per inputmean response time
16.6
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    69.9

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

  • Document understanding

    66.1

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

  • Information extraction

    67.6

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

  • Transcription

    88.0

    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 98.4±1.1 ≈1of 115 $74$0.37 run 17.69 T1405 · 17 Sep 2026prompt.txt
Blacklist Cards 81.3±3.5 ≈99of 114 $28$0.93 run 22.29 T1412 · 25 Jul 2026prompt.txt
Book Advert XML Repair 44.3±9.9 ≈16of 119 $37$1.86 run 13.89 T1415 · 25 Jul 2026prompt.txt
Business Letters 72.0±6.5 ≈1of 114 $43$2.45 run 4.74 T1406 · 17 Sep 2026prompt.txt
Company Lists 63.7±13.5 ≈97of 114 $57$0.85 run 21.89 T1413 · 25 Jul 2026prompt.txt
Duty Rosters 14.2±14.8 ≈101of 107 $43$0.91 run 15.73 T1419 · 17 Sep 2026prompt.txt
Fraktur Adverts 99.3±0.2 ≈1of 114 $92$0.46 run 43.44 T1409 · 17 Sep 2026prompt_optimized.txt
General Meeting Minutes 68.5±26.0 ≈85of 108 $53$0.48 run 12.8 T1418 · 17 Sep 2026prompt.txt
Library Cards 75.4±4.0 ≈98of 114 $22$5.70 run 8.37 T1410 · 25 Jul 2026prompt.txt
Magazine Pages 88.6±7.6 ≈18of 115 $26$1.19 run 16.39 T1417 · 25 Jul 2026prompt.txt
Medieval Manuscripts 69.1±20.1 ≈27of 114 $34$0.41 run 10.62 T1411 · 17 Sep 2026prompt.txt
Personnel Cards 95.5±3.0 ≈58of 114 $63$3.83 run 11.17 T1416 · 17 Sep 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
25 Jul 2026525archived price page

Every run

  1. T1419Duty Rosters17 Sep 202614.2$0.91
  2. T1418General Meeting Minutes17 Sep 202668.5$0.48
  3. T1417Magazine Pages17 Sep 202679.1$1.19
  4. T1416Personnel Cards17 Sep 202695.5$3.83
  5. T1415Book Advert XML Repair17 Sep 202636.7$1.89
  6. T1414Company Lists17 Sep 202652.1$0.81
  7. T1413Company Lists17 Sep 202663.5$0.84
  8. T1412Blacklist Cards17 Sep 202681.0$0.93
  9. T1411Medieval Manuscripts17 Sep 202669.1$0.41
  10. T1410Library Cards17 Sep 202634.2$5.57
  11. T1409Fraktur Adverts17 Sep 202699.3$0.46
  12. T1408Business Letters17 Sep 2026 · 32 signed letters70.0$1.53
  13. T1407Business Letters17 Sep 2026 · 25 unsigned letters86.0$0.92
  14. T1406Business Letters17 Sep 2026 · all 57 letters72.0$2.45
  15. T1405Bibliographic Data17 Sep 202698.4$0.37
  16. T1419Duty Rosters25 Jul 20264.3$0.97
  17. T1418General Meeting Minutes25 Jul 202635.8$0.47
  18. T1417Magazine Pages25 Jul 202688.6$1.19
  19. T1416Personnel Cards25 Jul 202691.6$3.76
  20. T1415Book Advert XML Repair25 Jul 202644.3$1.86
  21. T1414Company Lists25 Jul 202656.5$0.80
  22. T1413Company Lists25 Jul 202663.7$0.85
  23. T1412Blacklist Cards25 Jul 202681.3$0.93
  24. T1411Medieval Manuscripts25 Jul 202666.9$0.41
  25. T1410Library Cards25 Jul 202675.4$5.70
  26. T1409Fraktur Adverts25 Jul 202679.6$0.45
  27. T1408Business Letters25 Jul 2026 · 32 signed letters67.0$1.54
  28. T1407Business Letters25 Jul 2026 · 25 unsigned letters76.0$0.92
  29. T1406Business Letters25 Jul 2026 · all 57 letters69.0$2.46
  30. T1405Bibliographic Data25 Jul 202678.7$0.38