University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

qwen/qwen3.5-flash-02-23

Qwenvia OpenRouter
average±3.2 (95%)
69.0
leaderboard rankof 114
≈63
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$1.59
seconds per inputmean response time
87.8
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    57.0

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

  • Document understanding

    56.6

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

  • Information extraction

    81.6

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

  • Transcription

    80.3

    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 93.0±2.1 ≈57of 115 $1.08$0.005 run 21.95 T0970 · 21 Apr 2026prompt.txt
Blacklist Cards 93.3±3.1 ≈51of 114 $0.66$0.022 run 21.55 T0977 · 17 Sep 2026prompt.txt
Book Advert XML Repair 16.4±7.6 ≈89of 119 $12$0.62 run 838.04 T0980 · 21 Apr 2026prompt.txt
Business Letters 56.0±6.8 ≈75of 114 $0.34$0.020 run 3.53 T0971 · 21 Apr 2026prompt.txt
Company Lists 86.2±8.2 ≈29of 114 $0.48$0.007 run 8.6 T0978 · 21 Apr 2026prompt.txt
Duty Rosters 72.2±5.2 ≈64of 107 $1.42$0.030 run 97.02 T1328 · 22 Jul 2026prompt.txt
Fraktur Adverts 65.8±32.6 ≈67of 114 $0.75$0.004 run 13.45 T0974 · 21 Apr 2026prompt_optimized.txt
General Meeting Minutes 86.4±7.1 ≈25of 108 $0.70$0.006 run 29.57 T1267 · 3 Jul 2026prompt.txt
Library Cards 84.0±1.7 ≈73of 114 $0.17$0.044 run 3.7 T0975 · 21 Apr 2026prompt.txt
Magazine Pages 0.0±0.0 ≈79of 115 $0.21$0.010 run 3.05 T0982 · 21 Apr 2026prompt.txt
Medieval Manuscripts 77.6±12.0 ≈27of 114 $0.27$0.003 run 4.2 T0976 · 21 Apr 2026prompt.txt
Personnel Cards 97.5±1.0 ≈1of 114 $0.61$0.037 run 9.12 T0981 · 21 Apr 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
21 Apr 20260.0650.26archived price page

Every run

  1. T1267General Meeting Minutes18 Sep 202679.3$0.020
  2. T0982Magazine Pages17 Sep 20260.0$0.034
  3. T0981Personnel Cards17 Sep 202695.7$0.076
  4. T0980Book Advert XML Repair17 Sep 20260.0$0.001
  5. T0979Company Lists17 Sep 202670.6$0.016
  6. T0978Company Lists17 Sep 202679.5$0.016
  7. T0977Blacklist Cards17 Sep 202693.3$0.022
  8. T0976Medieval Manuscripts17 Sep 202668.2$0.022
  9. T0975Library Cards17 Sep 202675.8$0.32
  10. T0974Fraktur Adverts17 Sep 202651.1$0.014
  11. T0973Business Letters17 Sep 2026 · 32 signed letters55.0$0.030
  12. T0972Business Letters17 Sep 2026 · 25 unsigned letters49.0$0.024
  13. T0971Business Letters17 Sep 2026 · all 57 letters48.0$0.051
  14. T0970Bibliographic Data17 Sep 202674.8$0.011
  15. T1328Duty Rosters22 Jul 202672.2$0.030
  16. T1267General Meeting Minutes3 Jul 202686.4$0.006
  17. T0982Magazine Pages21 Apr 20260.0$0.010
  18. T0981Personnel Cards21 Apr 202697.5$0.037
  19. T0980Book Advert XML Repair21 Apr 202616.4$0.62
  20. T0979Company Lists21 Apr 202677.9$0.006
  21. T0978Company Lists21 Apr 202686.2$0.007
  22. T0977Blacklist Cards21 Apr 202693.1$0.008
  23. T0976Medieval Manuscripts21 Apr 202677.6$0.003
  24. T0975Library Cards21 Apr 202684.0$0.044
  25. T0974Fraktur Adverts21 Apr 202665.8$0.004
  26. T0973Business Letters21 Apr 2026 · 32 signed letters61.0$0.012
  27. T0972Business Letters21 Apr 2026 · 25 unsigned letters50.0$0.007
  28. T0971Business Letters21 Apr 2026 · all 57 letters56.0$0.020
  29. T0970Bibliographic Data21 Apr 202693.0$0.005