University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

gpt-6.1-sol

OpenAI
average±1.8 (95%)
87.5
leaderboard rankof 114
≈1
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$20
seconds per inputmean response time
19.8
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    81.5

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

  • Document understanding

    96.9

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

  • Information extraction

    86.9

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

  • Transcription

    93.1

    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 Dataother prompt 98.7±1.3 ≈1of 115 $39$0.20 run 25.58 T1914 · 24 Oct 2026prompt_v2.txt
Blacklist Cards 96.6±2.6 ≈8of 114 $8.71$0.29 run 4.95 T1921 · 24 Oct 2026prompt.txt
Book Advert XML Repairother prompt 65.0±10.5 ≈1of 119 $8.15$0.41 run 14.84 T1923 · 24 Oct 2026prompt_v2.txt
Business Letters 61.0±7.3 ≈24of 114 $15$0.85 run 7.15 T1915 · 24 Oct 2026prompt.txt
Company Listsother prompt 72.4±12.7 ≈80of 114 $18$0.26 run 11.64 T1922 · 24 Oct 2026prompt_v2.txt
Duty Rosters 96.0±1.5 ≈1of 107 $26$0.54 run 29.35 T1927 · 24 Oct 2026prompt.txt
Fraktur Adverts 97.1±0.4 ≈15of 114 $37$0.18 run 59.05 T1918 · 24 Oct 2026prompt_optimized.txt
General Meeting Minutes 92.1±5.6 ≈1of 108 $21$0.19 run 16.22 T1926 · 24 Oct 2026prompt.txt
Library Cardsother prompt 91.6±1.3 ≈1of 114 $6.25$1.64 run 5.45 T1919 · 24 Oct 2026prompt_year-string.txt
Magazine Pagesother prompt 96.8±2.8 ≈1of 115 $9.16$0.42 run 5.03 T1925 · 24 Oct 2026prompt_explicit-json.txt
Medieval Manuscripts 84.3±9.2 ≈1of 114 $28$0.34 run 47.27 T1920 · 24 Oct 2026prompt.txt
Personnel Cards 98.0±0.7 ≈1of 114 $21$1.28 run 10.83 T1924 · 24 Oct 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
24 Oct 20262.010.0archived price page

Every run

  1. T1927Duty Rosters24 Oct 202696.0$0.54
  2. T1926General Meeting Minutes24 Oct 202692.1$0.19
  3. T1925Magazine Pages24 Oct 202696.8$0.42
  4. T1924Personnel Cards24 Oct 202698.0$1.28
  5. T1923Book Advert XML Repair24 Oct 202665.0$0.41
  6. T1922Company Lists24 Oct 202672.4$0.26
  7. T1921Blacklist Cards24 Oct 202696.6$0.29
  8. T1920Medieval Manuscripts24 Oct 202684.3$0.34
  9. T1919Library Cards24 Oct 202691.6$1.64
  10. T1918Fraktur Adverts24 Oct 202697.1$0.18
  11. T1917Business Letters24 Oct 2026 · 32 signed letters65.0$0.55
  12. T1916Business Letters24 Oct 2026 · 25 unsigned letters59.0$0.31
  13. T1915Business Letters24 Oct 2026 · all 57 letters61.0$0.85
  14. T1914Bibliographic Data24 Oct 202698.7$0.20