University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

grok-4.6

xAI
average±1.6 (95%)
88.0
leaderboard rankof 114
≈1
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$44
seconds per inputmean response time
107.2
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    82.6

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

  • Document understanding

    86.0

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

  • Information extraction

    90.8

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

  • Transcription

    93.6

    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 98.2±1.7 ≈8of 115 $62$0.31 run 128.48 T1600 · 16 Sep 2026prompt.txt
Blacklist Cards 97.1±2.1 ≈1of 114 $13$0.41 run 17.87 T1607 · 18 Aug 2026prompt.txt
Book Advert XML Repair 67.4±10.5 ≈1of 119 $39$1.95 run 107.79 T1610 · 18 Aug 2026prompt.txt
Business Letters 70.0±6.8 ≈10of 114 $24$1.40 run 40.12 T1601 · 16 Sep 2026prompt.txt
Company Lists 98.0±1.4 ≈1of 114 $29$0.44 run 56.1 T1608 · 18 Aug 2026prompt.txt
Duty Rosters 88.3±2.5 ≈18of 107 $152$3.19 run 443.54 T1614 · 16 Sep 2026prompt.txt
Fraktur Adverts 98.6±0.4 ≈9of 114 $16$0.081 run 50.88 T1604 · 18 Aug 2026prompt_optimized.txt
General Meeting Minutes 92.8±6.0 ≈1of 108 $53$0.48 run 118.41 T1613 · 16 Sep 2026prompt.txt
Library Cards 91.0±1.3 ≈1of 114 $20$5.15 run 39.56 T1605 · 16 Sep 2026prompt.txt
Magazine Pages 71.8±10.9 ≈34of 115 $23$1.05 run 50.11 T1612 · 16 Sep 2026prompt.txt
Medieval Manuscripts 84.5±7.3 ≈1of 114 $84$1.01 run 220.55 T1606 · 16 Sep 2026prompt.txt
Personnel Cards 97.8±0.8 ≈1of 114 $15$0.94 run 12.74 T1611 · 18 Aug 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
18 Aug 20262.06.0archived price page

Every run

  1. T1614Duty Rosters16 Sep 202688.3$3.19
  2. T1613General Meeting Minutes16 Sep 202692.8$0.48
  3. T1612Magazine Pages16 Sep 202671.8$1.05
  4. T1611Personnel Cards16 Sep 202697.6$2.17
  5. T1610Book Advert XML Repair16 Sep 202642.3$2.48
  6. T1609Company Lists16 Sep 202677.0$0.41
  7. T1608Company Lists16 Sep 202684.1$0.53
  8. T1607Blacklist Cards16 Sep 202696.9$0.58
  9. T1606Medieval Manuscripts16 Sep 202684.5$1.01
  10. T1605Library Cards16 Sep 202691.0$5.15
  11. T1604Fraktur Adverts16 Sep 202679.0$0.36
  12. T1603Business Letters16 Sep 2026 · 32 signed letters71.0$0.88
  13. T1602Business Letters16 Sep 2026 · 25 unsigned letters74.0$0.49
  14. T1601Business Letters16 Sep 2026 · all 57 letters70.0$1.40
  15. T1600Bibliographic Data16 Sep 202698.2$0.31
  16. T1614Duty Rosters18 Aug 202687.0$2.23
  17. T1613General Meeting Minutes18 Aug 202692.4$0.34
  18. T1612Magazine Pages18 Aug 20262.3$0.29
  19. T1611Personnel Cards18 Aug 202697.8$0.94
  20. T1610Book Advert XML Repair18 Aug 202667.4$1.95
  21. T1609Company Lists18 Aug 202668.5$0.36
  22. T1608Company Lists18 Aug 202698.0$0.44
  23. T1607Blacklist Cards18 Aug 202697.1$0.41
  24. T1606Medieval Manuscripts18 Aug 202684.3$0.34
  25. T1605Library Cards18 Aug 202689.6$3.53
  26. T1604Fraktur Adverts18 Aug 202698.6$0.081
  27. T1603Business Letters18 Aug 2026 · 32 signed letters67.0$0.75
  28. T1602Business Letters18 Aug 2026 · 25 unsigned letters75.0$0.40
  29. T1601Business Letters18 Aug 2026 · all 57 letters66.0$1.11
  30. T1600Bibliographic Data18 Aug 202698.0$0.25