qwen/qwen3.8-flash
Qwenvia OpenRouter- average±2.3 (95%)
- 72.8
- leaderboard rankof 114
- ≈50
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $5.01
- seconds per inputmean response time
- 164.7
By task
-
Data correction
65.8
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
58.2
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
82.8
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
85.5
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 | 97.3±1.9 | ≈29of 115 | $6.26$0.031 run | 210.66 | T1720 · 8 Sep 2026prompt.txt |
| Blacklist Cards | 94.0±3.3 | ≈14of 114 | $1.33$0.044 run | 32.64 | T1727 · 18 Sep 2026prompt.txt |
| Book Advert XML Repair | 36.9±10.1 | ≈41of 119 | $6.13$0.31 run | 208.07 | T1730 · 8 Sep 2026prompt.txt |
| Business Letters | 55.0±6.5 | ≈84of 114 | $2.01$0.11 run | 98.48 | T1721 · 8 Sep 2026prompt.txt |
| Company Lists | 82.7±15.3 | ≈29of 114 | $1.86$0.028 run | 195.85 | T1728 · 8 Sep 2026prompt.txt |
| Duty Rosters | 80.0±2.3 | ≈46of 107 | $7.12$0.15 run | 217.24 | T1734 · 18 Sep 2026prompt.txt |
| Fraktur Adverts | 93.0±3.8 | ≈32of 114 | $13$0.066 run | 432.24 | T1724 · 8 Sep 2026prompt_optimized.txt |
| General Meeting Minutes | 85.1±6.8 | ≈25of 108 | $7.00$0.063 run | 163.6 | T1733 · 8 Sep 2026prompt.txt |
| Library Cards | 85.8±1.9 | ≈51of 114 | $2.64$0.70 run | 101.46 | T1725 · 18 Sep 2026prompt.txt |
| Magazine Pages | 0.0±0.0 | ≈79of 115 | $0.81$0.037 run | 12.31 | T1732 · 8 Sep 2026prompt.txt |
| Medieval Manuscripts | 68.8±16.0 | ≈68of 114 | $9.26$0.11 run | 257.21 | T1726 · 8 Sep 2026prompt.txt |
| Personnel Cards | 94.8±2.8 | ≈58of 114 | $2.56$0.16 run | 46.32 | T1731 · 8 Sep 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 8 Sep 2026 | 0.15 | 0.47 | archived price page |
Every run
- T1734Duty Rosters18 Sep 202680.0$0.15
- T1733General Meeting Minutes18 Sep 202678.8$0.064
- T1732Magazine Pages18 Sep 20260.0$0.032
- T1731Personnel Cards18 Sep 202692.5$0.15
- T1730Book Advert XML Repair18 Sep 202636.0$0.25
- T1729Company Lists18 Sep 202661.7$0.025
- T1728Company Lists18 Sep 202672.3$0.036
- T1727Blacklist Cards18 Sep 202694.0$0.044
- T1726Medieval Manuscripts18 Sep 202639.8$0.068
- T1725Library Cards18 Sep 202685.8$0.70
- T1724Fraktur Adverts18 Sep 202673.7$0.020
- T1723Business Letters18 Sep 2026 · 32 signed letters60.0$0.079
- T1722Business Letters18 Sep 2026 · 25 unsigned letters53.0$0.040
- T1721Business Letters18 Sep 2026 · all 57 letters54.0$0.11
- T1720Bibliographic Data18 Sep 202678.3$0.012
- T1734Duty Rosters8 Sep 202668.0$0.13
- T1733General Meeting Minutes8 Sep 202685.1$0.063
- T1732Magazine Pages8 Sep 20260.0$0.037
- T1731Personnel Cards8 Sep 202694.8$0.16
- T1730Book Advert XML Repair8 Sep 202636.9$0.31
- T1729Company Lists8 Sep 202664.4$0.018
- T1728Company Lists8 Sep 202682.7$0.028
- T1727Blacklist Cards8 Sep 202688.7$0.045
- T1726Medieval Manuscripts8 Sep 202668.8$0.11
- T1725Library Cards8 Sep 202681.9$0.46
- T1724Fraktur Adverts8 Sep 202693.0$0.066
- T1723Business Letters8 Sep 2026 · 32 signed letters58.0$0.078
- T1722Business Letters8 Sep 2026 · 25 unsigned letters54.0$0.036
- T1721Business Letters8 Sep 2026 · all 57 letters55.0$0.11
- T1720Bibliographic Data8 Sep 202697.3$0.031