claude-opus-4-5-20251101
Anthropic- average±1.9 (95%)
- 77.0
- leaderboard rankof 114
- ≈37
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $35
- seconds per inputmean response time
- 13.8
By task
-
Data correction
70.9
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
60.7
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
84.5
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
94.1
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 | 97.0±1.8 | ≈29of 115 | $54$0.27 run | 20.62 | T0510 · 16 Sep 2026prompt.txt |
| Blacklist Cards | 97.0±2.3 | ≈1of 114 | $19$0.63 run | 5.79 | T0517 · 24 Jan 2026prompt.txt |
| Book Advert XML Repair | 45.0±9.6 | ≈16of 119 | $25$1.27 run | 9.31 | T0520 · 24 Jan 2026prompt.txt |
| Business Letters | 61.0±7.3 | ≈24of 114 | $22$1.24 run | 4.82 | T0511 · 24 Jan 2026prompt.txt |
| Company Lists | 85.2±12.7 | ≈29of 114 | $36$0.54 run | 13.46 | T0518 · 24 Jan 2026prompt.txt |
| Duty Rosters | 79.8±3.3 | ≈46of 107 | $61$1.27 run | 14.44 | T1278 · 17 Sep 2026prompt.txt |
| Fraktur Adverts | 97.8±1.5 | ≈9of 114 | $73$0.37 run | 53.56 | T0514 · 16 Sep 2026prompt_optimized.txt |
| General Meeting Minutes | 85.3±9.1 | ≈25of 108 | $36$0.32 run | 14.36 | T1212 · 3 Jul 2026prompt.txt |
| Library Cards | 85.8±1.4 | ≈51of 114 | $19$4.93 run | 5.62 | T0515 · 24 Jan 2026prompt.txt |
| Magazine Pages | 5.4±6.3 | ≈79of 115 | $14$0.65 run | 3.39 | T0805 · 23 Mar 2026prompt.txt |
| Medieval Manuscripts | 87.6±7.3 | ≈1of 114 | $22$0.26 run | 11.46 | T0516 · 24 Jan 2026prompt.txt |
| Personnel Cards | 96.8±1.9 | ≈30of 114 | $40$2.43 run | 8.24 | T0579 · 16 Sep 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 23 Jan 2026 | 5.0 | 25.0 | archived price page |
Every run
- T1278Duty Rosters17 Sep 202679.8$1.27
- T1212General Meeting Minutes17 Sep 202684.6$0.32
- T0513Business Letters17 Sep 2026 · 32 signed letters64.0$0.76
- T0512Business Letters17 Sep 2026 · 25 unsigned letters55.0$0.48
- T0511Business Letters17 Sep 2026 · all 57 letters60.0$1.24
- T0805Magazine Pages16 Sep 20265.4$0.65
- T0579Personnel Cards16 Sep 202696.8$2.43
- T0520Book Advert XML Repair16 Sep 202644.4$1.26
- T0519Company Lists16 Sep 202673.1$0.52
- T0518Company Lists16 Sep 202682.5$0.54
- T0517Blacklist Cards16 Sep 202696.4$0.63
- T0516Medieval Manuscripts16 Sep 202684.6$0.26
- T0515Library Cards16 Sep 202685.6$4.93
- T0514Fraktur Adverts16 Sep 202697.8$0.37
- T0513Business Letters16 Sep 2026 · 32 signed letters62.0$0.76
- T0512Business Letters16 Sep 2026 · 25 unsigned letters55.0$0.48
- T0511Business Letters16 Sep 2026 · all 57 letters59.0$1.24
- T0510Bibliographic Data16 Sep 202697.0$0.27
- T1278Duty Rosters24 Jul 202678.2$1.27
- T1212General Meeting Minutes3 Jul 202685.3$0.32
- T0805Magazine Pages23 Mar 20265.4$0.65
- T0579Personnel Cards9 Feb 202691.2$2.37
- T0520Book Advert XML Repair24 Jan 202645.0$1.27
- T0519Company Lists24 Jan 202670.9$0.52
- T0518Company Lists24 Jan 202685.2$0.54
- T0517Blacklist Cards24 Jan 202697.0$0.63
- T0516Medieval Manuscripts24 Jan 202687.6$0.26
- T0515Library Cards24 Jan 202685.8$4.93
- T0514Fraktur Adverts24 Jan 202638.0$0.18
- T0513Business Letters24 Jan 2026 · 32 signed letters64.0$0.76
- T0512Business Letters24 Jan 2026 · 25 unsigned letters56.0$0.48
- T0511Business Letters24 Jan 2026 · all 57 letters61.0$1.24
- T0510Bibliographic Data24 Jan 202696.8$0.27