qwen/qwen3.8-max-prime
Qwenvia OpenRouter- average±1.9 (95%)
- 74.3
- leaderboard rankof 114
- ≈44
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $87
- seconds per inputmean response time
- 89.4
By task
-
Data correction
63.6
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
60.3
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
91.3
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 Dataother prompt | 97.6±1.9 | ≈8of 115 | $122$0.61 run | 129.9 | T2068 · 25 Oct 2026prompt_v2.txt |
| Blacklist Cards | 93.5±2.6 | ≈45of 114 | $49$1.61 run | 44.84 | T2075 · 25 Oct 2026prompt.txt |
| Book Advert XML Repairother prompt | 29.3±9.7 | ≈41of 119 | $121$6.05 run | 135.99 | T2077 · 25 Oct 2026prompt_v2.txt |
| Business Letters | 57.0±7.2 | ≈65of 114 | $54$3.08 run | 57.74 | T2069 · 25 Oct 2026prompt.txt |
| Company Listsother prompt | 82.2±10.3 | ≈55of 114 | $69$1.04 run | 65.04 | T2076 · 25 Oct 2026prompt_v2.txt |
| Duty Rosters | 83.1±4.4 | ≈30of 107 | $145$3.04 run | 159.79 | T2081 · 25 Oct 2026prompt.txt |
| Fraktur Adverts | 94.6±4.0 | ≈32of 114 | $132$0.66 run | 135.32 | T2072 · 25 Oct 2026prompt_optimized.txt |
| General Meeting Minutes | 85.8±8.0 | ≈25of 108 | $106$0.95 run | 117.65 | T2080 · 25 Oct 2026prompt.txt |
| Library Cardsother prompt | 88.6±1.7 | ≈15of 114 | $40$10 run | 45.2 | T2073 · 25 Oct 2026prompt_year-string.txt |
| Magazine Pagesother prompt | 0.0±0.0 | ≈79of 115 | $39$1.80 run | 32.62 | T2079 · 25 Oct 2026prompt_explicit-json.txt |
| Medieval Manuscripts | 81.6±12.0 | ≈27of 114 | $110$1.32 run | 101.62 | T2074 · 25 Oct 2026prompt.txt |
| Personnel Cards | 97.8±1.0 | ≈1of 114 | $53$3.22 run | 46.86 | T2078 · 25 Oct 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 24 Oct 2026 | 4.0 | 12.0 | archived price page |
Every run
- T2081Duty Rosters25 Oct 202683.1$3.04
- T2080General Meeting Minutes25 Oct 202685.8$0.95
- T2079Magazine Pages25 Oct 20260.0$1.80
- T2078Personnel Cards25 Oct 202697.8$3.22
- T2077Book Advert XML Repair25 Oct 202629.3$6.05
- T2076Company Lists25 Oct 202682.2$1.04
- T2075Blacklist Cards25 Oct 202693.5$1.61
- T2074Medieval Manuscripts25 Oct 202681.6$1.32
- T2073Library Cards25 Oct 202688.6$10
- T2072Fraktur Adverts25 Oct 202694.6$0.66
- T2071Business Letters25 Oct 2026 · 32 signed letters61.0$1.49
- T2070Business Letters25 Oct 2026 · 25 unsigned letters56.0$1.16
- T2069Business Letters25 Oct 2026 · all 57 letters57.0$3.08
- T2068Bibliographic Data25 Oct 202697.6$0.61