meta/muse-spark-open
Metavia OpenRouter- average±2.6 (95%)
- 75.8
- leaderboard rankof 114
- ≈37
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $14
- seconds per inputmean response time
- 22.5
By task
-
Data correction
62.6
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
65.2
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
84.0
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
92.4
Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.
By benchmark
| Benchmark | ScoreThe 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 docs | s / input | Best run |
|---|---|---|---|---|---|
| Bibliographic Dataother prompt | 84.2±24.2 | ≈74of 115 | $26$0.13 run | 60.45 | T2026 · 24 Oct 2026prompt_v2.txt |
| Blacklist Cards | 95.9±2.6 | ≈14of 114 | $5.45$0.18 run | 4.74 | T2033 · 24 Oct 2026prompt.txt |
| Book Advert XML Repairother prompt | 26.7±7.6 | ≈64of 119 | $18$0.89 run | 26.43 | T2035 · 24 Oct 2026prompt_v2.txt |
| Business Letters | 66.0±6.9 | ≈10of 114 | $10$0.60 run | 5.88 | T2027 · 24 Oct 2026prompt.txt |
| Company Listsother prompt | 86.1±6.6 | ≈29of 114 | $16$0.24 run | 17.74 | T2034 · 24 Oct 2026prompt_v2.txt |
| Duty Rosters | 79.8±3.5 | ≈46of 107 | $19$0.41 run | 40.52 | T2039 · 24 Oct 2026prompt.txt |
| Fraktur Adverts | 95.6±3.9 | ≈32of 114 | $20$0.10 run | 40.4 | T2030 · 24 Oct 2026prompt_optimized.txt |
| General Meeting Minutes | 86.0±6.9 | ≈25of 108 | $15$0.13 run | 28.39 | T2038 · 24 Oct 2026prompt.txt |
| Library Cardsother prompt | 89.7±1.4 | ≈15of 114 | $6.20$1.63 run | 22.39 | T2031 · 24 Oct 2026prompt_year-string.txt |
| Magazine Pagesother prompt | 17.5±10.9 | ≈66of 115 | $5.95$0.27 run | 3.61 | T2037 · 24 Oct 2026prompt_explicit-json.txt |
| Medieval Manuscripts | 83.3±7.8 | ≈1of 114 | $17$0.20 run | 13.5 | T2032 · 24 Oct 2026prompt.txt |
| Personnel Cards | 98.4±0.8 | ≈1of 114 | $8.70$0.53 run | 5.73 | T2036 · 24 Oct 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 24 Oct 2026 | 1.25 | 4.25 | archived price page |
Every run
- T2039Duty Rosters24 Oct 202679.8$0.41
- T2038General Meeting Minutes24 Oct 202686.0$0.13
- T2037Magazine Pages24 Oct 202617.5$0.27
- T2036Personnel Cards24 Oct 202698.4$0.53
- T2035Book Advert XML Repair24 Oct 202626.7$0.89
- T2034Company Lists24 Oct 202686.1$0.24
- T2033Blacklist Cards24 Oct 202695.9$0.18
- T2032Medieval Manuscripts24 Oct 202683.3$0.20
- T2031Library Cards24 Oct 202689.7$1.63
- T2030Fraktur Adverts24 Oct 202695.6$0.10
- T2029Business Letters24 Oct 2026 · 32 signed letters63.0$0.35
- T2028Business Letters24 Oct 2026 · 25 unsigned letters67.0$0.16
- T2027Business Letters24 Oct 2026 · all 57 letters66.0$0.60
- T2026Bibliographic Data24 Oct 202684.2$0.13