University of Basel RISE · Research and Infrastructure SupportHumanities Data Benchmark

meta/muse-spark-1.3

Metavia OpenRouter
average±3.6 (95%)
83.8
leaderboard rankof 114
≈14
benchmarksof 12
12
per 1,000 documentsmean over its benchmarks
$21
seconds per inputmean response time
31.5
Compare with other models Its row in the Matrix All its test runs

By task

  • Data correction

    61.2

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

  • Document understanding

    94.7

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

  • Information extraction

    89.0

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

  • Transcription

    86.9

    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 98.0±1.5 ≈8of 115 $32$0.16 run 40.11 T1690 · 17 Sep 2026prompt.txt
Blacklist Cards 97.4±1.5 ≈1of 114 $9.98$0.33 run 10.48 T1697 · 17 Sep 2026prompt.txt
Book Advert XML Repair 24.2±7.7 ≈64of 119 $26$1.28 run 44.55 T1700 · 17 Sep 2026prompt.txt
Business Letters 68.0±6.5 ≈10of 114 $16$0.91 run 17.97 T1691 · 8 Sep 2026prompt.txt
Company Lists 93.6±4.2 ≈6of 114 $18$0.28 run 19.33 T1698 · 17 Sep 2026prompt.txt
Duty Rosters 87.9±3.1 ≈18of 107 $43$0.90 run 81.66 T1704 · 18 Sep 2026prompt.txt
Fraktur Adverts 78.8±38.6 ≈32of 114 $30$0.15 run 50.13 T1694 · 17 Sep 2026prompt_optimized.txt
General Meeting Minutes 87.6±14.9 ≈25of 108 $25$0.23 run 38.07 T1703 · 17 Sep 2026prompt.txt
Library Cards 90.6±1.5 ≈1of 114 $12$3.25 run 19.9 T1695 · 8 Sep 2026prompt.txt
Magazine Pages 97.9±2.4 ≈1of 115 $10$0.47 run 15.07 T1702 · 8 Sep 2026prompt.txt
Medieval Manuscripts 83.6±7.9 ≈1of 114 $16$0.19 run 22.4 T1696 · 17 Sep 2026prompt.txt
Personnel Cards 98.2±0.8 ≈1of 114 $17$1.03 run 17.96 T1701 · 17 Sep 2026prompt.txt

Price

FromInput, $ per million tokensOutput, $ per million tokensSource
8 Sep 20261.254.25archived price page

Every run

  1. T1704Duty Rosters18 Sep 202687.9$0.90
  2. T1703General Meeting Minutes17 Sep 202687.6$0.23
  3. T1702Magazine Pages17 Sep 202695.8$0.48
  4. T1701Personnel Cards17 Sep 202698.2$1.03
  5. T1700Book Advert XML Repair17 Sep 202624.2$1.28
  6. T1699Company Lists17 Sep 202681.7$0.24
  7. T1698Company Lists17 Sep 202693.6$0.28
  8. T1697Blacklist Cards17 Sep 202697.4$0.33
  9. T1696Medieval Manuscripts17 Sep 202683.6$0.19
  10. T1695Library Cards17 Sep 202690.5$4.28
  11. T1694Fraktur Adverts17 Sep 202678.8$0.15
  12. T1693Business Letters17 Sep 2026 · 32 signed letters66.0$0.55
  13. T1692Business Letters17 Sep 2026 · 25 unsigned letters70.0$0.34
  14. T1691Business Letters17 Sep 2026 · all 57 letters67.0$0.92
  15. T1690Bibliographic Data17 Sep 202698.0$0.16
  16. T1704Duty Rosters8 Sep 202686.7$0.82
  17. T1703General Meeting Minutes8 Sep 202684.5$0.24
  18. T1702Magazine Pages8 Sep 202697.9$0.47
  19. T1701Personnel Cards8 Sep 202698.2$1.02
  20. T1700Book Advert XML Repair8 Sep 202621.8$1.21
  21. T1699Company Lists8 Sep 202667.7$0.24
  22. T1698Company Lists8 Sep 202683.9$0.28
  23. T1697Blacklist Cards8 Sep 202694.6$0.32
  24. T1696Medieval Manuscripts8 Sep 202679.9$0.20
  25. T1695Library Cards8 Sep 202690.6$3.25
  26. T1694Fraktur Adverts8 Sep 202677.6$0.15
  27. T1693Business Letters8 Sep 2026 · 32 signed letters67.0$0.57
  28. T1692Business Letters8 Sep 2026 · 25 unsigned letters75.0$0.33
  29. T1691Business Letters8 Sep 2026 · all 57 letters68.0$0.91
  30. T1690Bibliographic Data8 Sep 202697.1$0.16