qwen/qwen3.8-27b
Qwenvia OpenRouter- average±2.3 (95%)
- 80.7
- leaderboard rankof 114
- ≈25
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $15
- seconds per inputmean response time
- 327.9
By task
-
Data correction
74.5
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
84.2
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
84.2
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
84.2
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 | 94.1±5.9 | ≈57of 115 | $25$0.12 run | 249.19 | T1735 · 18 Sep 2026prompt.txt |
| Blacklist Cards | 95.3±3.0 | ≈14of 114 | $1.73$0.057 run | 16.09 | T1742 · 18 Sep 2026prompt.txt |
| Book Advert XML Repair | 51.8±10.5 | ≈16of 119 | $25$1.23 run | 360.81 | T1745 · 18 Sep 2026prompt.txt |
| Business Letters | 60.0±7.0 | ≈24of 114 | $19$1.09 run | 111.15 | T1736 · 18 Sep 2026prompt.txt |
| Company Lists | 90.9±3.5 | ≈6of 114 | $7.79$0.12 run | 52.88 | T1743 · 9 Sep 2026prompt.txt |
| Duty Rosters | 81.4±3.1 | ≈30of 107 | $37$0.78 run | 344.22 | T1749 · 9 Sep 2026prompt.txt |
| Fraktur Adverts | 82.7±14.0 | ≈32of 114 | $13$0.067 run | 416.38 | T1739 · 18 Sep 2026prompt_optimized.txt |
| General Meeting Minutes | 84.0±8.2 | ≈25of 108 | $13$0.11 run | 350.1 | T1748 · 9 Sep 2026prompt.txt |
| Library Cards | 84.0±2.3 | ≈73of 114 | $15$3.91 run | 1814.73 | T1740 · 9 Sep 2026prompt.txt |
| Magazine Pages | 74.1±12.3 | ≈34of 115 | $4.35$0.20 run | 27.83 | T1747 · 18 Sep 2026prompt.txt |
| Medieval Manuscripts | 72.7±10.4 | ≈27of 114 | $14$0.17 run | 127.45 | T1741 · 18 Sep 2026prompt.txt |
| Personnel Cards | 97.2±0.9 | ≈30of 114 | $9.32$0.57 run | 63.53 | T1746 · 18 Sep 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 18 Sep 2026 | 0.214 | 2.55 | archived price page |
| 8 Sep 2026 | 0.42 | 3.0 | archived price page |
Every run
- T1749Duty Rosters18 Sep 202676.6$0.67
- T1748General Meeting Minutes18 Sep 202683.8$0.14
- T1747Magazine Pages18 Sep 202674.1$0.20
- T1746Personnel Cards18 Sep 202697.2$0.57
- T1745Book Advert XML Repair18 Sep 202651.8$1.23
- T1744Company Lists18 Sep 202655.5$0.073
- T1743Company Lists18 Sep 202685.5$0.093
- T1742Blacklist Cards18 Sep 202695.3$0.057
- T1741Medieval Manuscripts18 Sep 202672.7$0.17
- T1740Library Cards18 Sep 202682.1$2.26
- T1739Fraktur Adverts18 Sep 202682.7$0.067
- T1738Business Letters18 Sep 2026 · 32 signed letters60.0$0.57
- T1737Business Letters18 Sep 2026 · 25 unsigned letters59.0$0.44
- T1736Business Letters18 Sep 2026 · all 57 letters60.0$1.09
- T1735Bibliographic Data18 Sep 202694.1$0.12
- T1749Duty Rosters9 Sep 202681.4$0.78
- T1748General Meeting Minutes9 Sep 202684.0$0.11
- T1747Magazine Pages9 Sep 202661.3$0.20
- T1746Personnel Cards9 Sep 202697.1$0.60
- T1745Book Advert XML Repair9 Sep 202649.5$1.87
- T1744Company Lists9 Sep 202645.3$0.12
- T1743Company Lists9 Sep 202690.9$0.12
- T1742Blacklist Cards9 Sep 202692.6$0.098
- T1741Medieval Manuscripts9 Sep 202672.0$0.22
- T1740Library Cards9 Sep 202684.0$3.91
- T1739Fraktur Adverts8 Sep 202644.4$0.27
- T1738Business Letters8 Sep 2026 · 32 signed letters61.0$0.63
- T1737Business Letters8 Sep 2026 · 25 unsigned letters57.0$0.38
- T1736Business Letters8 Sep 2026 · all 57 letters58.0$1.12
- T1735Bibliographic Data8 Sep 202685.9$0.090