University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

gpt-6-luna

OpenAI
average±1.9 (95%)
81.8
leaderboard rankof 114
≈25
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$1.28
seconds per inputmean response time
14.4
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    71.6

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

  • Document understanding

    94.0

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

  • Information extraction

    85.2

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

  • Transcription

    80.9

    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 97.2±1.8 ≈29of 115 $2.31$0.012 run 22.7 T1840 · 27 Sep 2026prompt.txt
Blacklist Cards 91.0±4.0 ≈51of 114 $0.55$0.018 run 4.7 T1847 · 27 Sep 2026prompt.txt
Book Advert XML Repair 46.5±9.6 ≈16of 119 $0.58$0.029 run 7.91 T1850 · 27 Sep 2026prompt.txt
Business Letters 62.0±7.1 ≈24of 114 $0.82$0.047 run 4.79 T1841 · 27 Sep 2026prompt.txt
Company Lists 83.5±8.8 ≈29of 114 $0.99$0.015 run 9.16 T1848 · 27 Sep 2026prompt.txt
Duty Rosters 88.5±2.5 ≈18of 107 $2.41$0.051 run 30.32 T1854 · 27 Sep 2026prompt.txt
Fraktur Adverts 95.3±3.6 ≈32of 114 $2.61$0.013 run 36.17 T1844 · 27 Sep 2026prompt_optimized.txt
General Meeting Minutes 87.6±7.9 ≈25of 108 $1.41$0.013 run 15.63 T1853 · 27 Sep 2026prompt.txt
Library Cardsother prompt 86.5±1.4 ≈41of 114 $0.44$0.12 run 5.75 T1845 · 27 Sep 2026prompt_year-string.txt
Magazine Pagesother prompt 96.6±2.8 ≈1of 115 $0.58$0.027 run 4.95 T1852 · 27 Sep 2026prompt_explicit-json.txt
Medieval Manuscripts 50.7±13.4 ≈95of 114 $1.44$0.017 run 20.67 T1846 · 27 Sep 2026prompt.txt
Personnel Cards 96.7±1.1 ≈30of 114 $1.19$0.073 run 9.71 T1851 · 27 Sep 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
27 Sep 20260.10.5archived price page

Every run

  1. T1854Duty Rosters27 Sep 202688.5$0.051
  2. T1853General Meeting Minutes27 Sep 202687.6$0.013
  3. T1852Magazine Pages27 Sep 202696.6$0.027
  4. T1851Personnel Cards27 Sep 202696.7$0.073
  5. T1850Book Advert XML Repair27 Sep 202646.5$0.029
  6. T1849Company Lists27 Sep 202675.1$0.014
  7. T1848Company Lists27 Sep 202683.5$0.015
  8. T1847Blacklist Cards27 Sep 202691.0$0.018
  9. T1846Medieval Manuscripts27 Sep 202650.7$0.017
  10. T1845Library Cards27 Sep 202686.5$0.12
  11. T1844Fraktur Adverts27 Sep 202695.3$0.013
  12. T1843Business Letters27 Sep 2026 · 32 signed letters65.0$0.029
  13. T1842Business Letters27 Sep 2026 · 25 unsigned letters58.0$0.017
  14. T1841Business Letters27 Sep 2026 · all 57 letters62.0$0.047
  15. T1840Bibliographic Data27 Sep 202697.2$0.012