meta-models/Muse-Glimmer-30B:together
Metavia Hugging Face- average±2.9 (95%)
- 77.4
- leaderboard rankof 114
- ≈37
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $7.35
- seconds per inputmean response time
- 102.2
By task
-
Data correction
69.9
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
72.0
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
85.6
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
81.3
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 Data | 88.2±15.1 | ≈74of 115 | $8.32$0.042 run | 97.01 | T1570 · 15 Aug 2026prompt.txt |
| Blacklist Cards | 96.0±2.1 | ≈14of 114 | $2.54$0.084 run | 10.87 | T1577 · 16 Aug 2026prompt.txt |
| Book Advert XML Repair | 43.1±11.1 | ≈16of 119 | $9.24$0.46 run | 66.03 | T1580 · 16 Aug 2026prompt.txt |
| Business Letters | 71.0±7.0 | ≈7of 114 | $4.44$0.25 run | 59.76 | T1571 · 15 Aug 2026prompt.txt |
| Company Lists | 92.3±4.7 | ≈6of 114 | $6.82$0.10 run | 39.02 | T1578 · 16 Aug 2026prompt.txt |
| Duty Rosters | 76.2±5.9 | ≈46of 107 | $16$0.33 run | 203.56 | T1584 · 16 Aug 2026prompt.txt |
| Fraktur Adverts | 83.5±15.3 | ≈32of 114 | $13$0.065 run | 176.01 | T1574 · 15 Aug 2026prompt_optimized.txt |
| General Meeting Minutes | 86.0±9.5 | ≈25of 108 | $5.70$0.051 run | 30.52 | T1583 · 16 Aug 2026prompt.txt |
| Library Cards | 89.3±1.4 | ≈15of 114 | $4.46$1.17 run | 386.66 | T1575 · 15 Aug 2026prompt.txt |
| Magazine Pages | 43.2±14.1 | ≈47of 115 | $3.25$0.15 run | 50.22 | T1582 · 18 Sep 2026prompt.txt |
| Medieval Manuscripts | 63.9±13.6 | ≈72of 114 | $8.58$0.10 run | 68.73 | T1576 · 16 Aug 2026prompt.txt |
| Personnel Cards | 96.6±1.2 | ≈30of 114 | $6.06$0.37 run | 38.32 | T1581 · 16 Aug 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 15 Aug 2026 | 0.35 | 1.5 | archived price page |
Every run
- T1583General Meeting Minutes18 Sep 202683.3$0.052
- T1582Magazine Pages18 Sep 202643.2$0.15
- T1581Personnel Cards18 Sep 2026––
- T1579Company Lists18 Sep 202638.9$0.060
- T1578Company Lists18 Sep 202691.7$0.091
- T1577Blacklist Cards18 Sep 202695.2$0.085
- T1573Business Letters18 Sep 2026 · 32 signed letters70.0$0.15
- T1572Business Letters18 Sep 2026 · 25 unsigned letters83.0$0.097
- T1571Business Letters18 Sep 2026 · all 57 letters71.0$0.25
- T1584Duty Rosters16 Aug 202676.2$0.33
- T1583General Meeting Minutes16 Aug 202686.0$0.051
- T1582Magazine Pages16 Aug 202639.6$0.14
- T1581Personnel Cards16 Aug 202696.6$0.37
- T1580Book Advert XML Repair16 Aug 202643.1$0.46
- T1579Company Lists16 Aug 202639.1$0.059
- T1578Company Lists16 Aug 202692.3$0.10
- T1577Blacklist Cards16 Aug 202696.0$0.084
- T1576Medieval Manuscripts16 Aug 202663.9$0.10
- T1575Library Cards15 Aug 202689.3$1.17
- T1574Fraktur Adverts15 Aug 202683.5$0.065
- T1573Business Letters15 Aug 2026 · 32 signed letters69.0$0.16
- T1572Business Letters15 Aug 2026 · 25 unsigned letters85.0$0.096
- T1571Business Letters15 Aug 2026 · all 57 letters71.0$0.25
- T1570Bibliographic Data15 Aug 202688.2$0.042