University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

grok-4.3

xAI
average±2.8 (95%)
79.5
leaderboard rankof 114
≈25
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$7.94
seconds per inputmean response time
69.2
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    69.6

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

  • Document understanding

    77.9

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

  • Information extraction

    85.8

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

  • Transcription

    82.8

    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 96.7±1.9 ≈40of 115 $14$0.069 run 28.17 T1102 · 8 Jun 2026prompt.txt
Blacklist Cards 96.6±2.3 ≈8of 114 $5.82$0.19 run 12.64 T1109 · 8 Jun 2026prompt.txt
Book Advert XML Repair 43.9±10.5 ≈16of 119 $4.77$0.24 run 13.34 T1112 · 8 Jun 2026prompt.txt
Business Letters 68.0±6.7 ≈10of 114 $8.03$0.46 run 8.45 T1103 · 16 Sep 2026prompt.txt
Company Lists 88.9±6.9 ≈6of 114 $8.41$0.13 run 22.96 T1110 · 8 Jun 2026prompt.txt
Duty Rosters 76.4±5.0 ≈46of 107 $10$0.22 run 26.52 T1338 · 16 Sep 2026prompt.txt
Fraktur Adverts 84.4±22.1 ≈32of 114 $9.16$0.046 run 32.79 T1106 · 8 Jun 2026prompt_optimized.txt
General Meeting Minutes 89.2±7.7 ≈1of 108 $8.27$0.074 run 15.7 T1275 · 16 Sep 2026prompt.txt
Library Cards 84.7±1.6 ≈51of 114 $4.47$1.18 run 631.65 T1107 · 8 Jun 2026prompt.txt
Magazine Pages 61.8±13.0 ≈34of 115 $4.97$0.23 run 11.15 T1114 · 8 Jun 2026prompt.txt
Medieval Manuscripts 68.6±12.3 ≈68of 114 $7.00$0.084 run 11.9 T1108 · 16 Sep 2026prompt.txt
Personnel Cards 95.4±2.3 ≈58of 114 $10$0.62 run 15.36 T1113 · 8 Jun 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
5 Jun 20261.252.5archived price page

Every run

  1. T1338Duty Rosters16 Sep 202676.4$0.22
  2. T1275General Meeting Minutes16 Sep 202689.2$0.074
  3. T1114Magazine Pages16 Sep 202657.2$0.24
  4. T1113Personnel Cards16 Sep 202695.2$0.62
  5. T1112Book Advert XML Repair16 Sep 202639.9$0.25
  6. T1111Company Lists16 Sep 202655.3$0.12
  7. T1110Company Lists16 Sep 202679.3$0.13
  8. T1109Blacklist Cards16 Sep 202696.5$0.20
  9. T1108Medieval Manuscripts16 Sep 202668.6$0.084
  10. T1107Library Cards16 Sep 202683.1$1.21
  11. T1106Fraktur Adverts16 Sep 202684.1$0.049
  12. T1105Business Letters16 Sep 2026 · 32 signed letters69.0$0.29
  13. T1104Business Letters16 Sep 2026 · 25 unsigned letters81.0$0.17
  14. T1103Business Letters16 Sep 2026 · all 57 letters68.0$0.46
  15. T1102Bibliographic Data16 Sep 202695.1$0.068
  16. T1338Duty Rosters24 Jul 202672.8$0.22
  17. T1275General Meeting Minutes3 Jul 202686.7$0.078
  18. T1114Magazine Pages8 Jun 202661.8$0.23
  19. T1113Personnel Cards8 Jun 202695.4$0.62
  20. T1112Book Advert XML Repair8 Jun 202643.9$0.24
  21. T1111Company Lists8 Jun 202663.6$0.12
  22. T1110Company Lists8 Jun 202688.9$0.13
  23. T1109Blacklist Cards8 Jun 202696.6$0.19
  24. T1108Medieval Manuscripts8 Jun 202668.4$0.081
  25. T1107Library Cards8 Jun 202684.7$1.18
  26. T1106Fraktur Adverts8 Jun 202684.4$0.046
  27. T1105Business Letters8 Jun 2026 · 32 signed letters69.0$0.29
  28. T1104Business Letters8 Jun 2026 · 25 unsigned letters77.0$0.17
  29. T1103Business Letters8 Jun 2026 · all 57 letters66.0$0.45
  30. T1102Bibliographic Data8 Jun 202696.7$0.069