University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

qwen/qwen3.8-max-0902

Qwenvia OpenRouter
average±2.0 (95%)
74.8
leaderboard rankof 114
≈44
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$44
seconds per inputmean response time
136.2
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    67.1

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

  • Document understanding

    59.8

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

  • Information extraction

    83.7

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

  • Transcription

    91.5

    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 97.9±1.5 ≈8of 115 $58$0.29 run 211.2 T1780 · 18 Sep 2026prompt.txt
Blacklist Cards 93.7±2.6 ≈45of 114 $26$0.85 run 68.29 T1787 · 18 Sep 2026prompt.txt
Book Advert XML Repair 36.9±9.8 ≈41of 119 $58$2.92 run 277.86 T1790 · 18 Sep 2026prompt.txt
Business Letters 57.0±6.3 ≈65of 114 $31$1.76 run 71.07 T1781 · 18 Sep 2026prompt.txt
Company Lists 80.7±14.1 ≈58of 114 $38$0.56 run 92.93 T1788 · 18 Sep 2026prompt.txt
Duty Rosters 81.9±4.5 ≈30of 107 $81$1.70 run 253.01 T1794 · 18 Sep 2026prompt.txt
Fraktur Adverts 94.9±4.1 ≈32of 114 $61$0.30 run 171.75 T1784 · 18 Sep 2026prompt_optimized.txt
General Meeting Minutes 86.0±8.1 ≈25of 108 $49$0.44 run 136.89 T1793 · 18 Sep 2026prompt.txt
Library Cards 88.6±1.7 ≈15of 114 $28$7.32 run 84.46 T1785 · 18 Sep 2026prompt.txt
Magazine Pages 0.0±0.0 ≈79of 115 $20$0.92 run 47.14 T1792 · 18 Sep 2026prompt.txt
Medieval Manuscripts 82.3±11.9 ≈1of 114 $51$0.61 run 156.96 T1786 · 18 Sep 2026prompt.txt
Personnel Cards 97.4±1.0 ≈1of 114 $30$1.84 run 62.98 T1791 · 18 Sep 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
16 Sep 20262.06.0archived price page

Every run

  1. T1794Duty Rosters18 Sep 202681.9$1.70
  2. T1793General Meeting Minutes18 Sep 202686.0$0.44
  3. T1792Magazine Pages18 Sep 20260.0$0.92
  4. T1791Personnel Cards18 Sep 202697.4$1.84
  5. T1790Book Advert XML Repair18 Sep 202636.9$2.92
  6. T1789Company Lists18 Sep 202674.5$0.34
  7. T1788Company Lists18 Sep 202680.7$0.56
  8. T1787Blacklist Cards18 Sep 202693.7$0.85
  9. T1786Medieval Manuscripts18 Sep 202682.3$0.61
  10. T1785Library Cards18 Sep 202688.6$7.32
  11. T1784Fraktur Adverts18 Sep 202694.9$0.30
  12. T1783Business Letters18 Sep 2026 · 32 signed letters56.0$0.88
  13. T1782Business Letters18 Sep 2026 · 25 unsigned letters58.0$0.64
  14. T1781Business Letters18 Sep 2026 · all 57 letters57.0$1.76
  15. T1780Bibliographic Data18 Sep 202697.9$0.29