minimax/minimax-m3.1-flash
MiniMaxvia OpenRouter- average±4.7 (95%)
- 65.0
- leaderboard rankof 114
- ≈79
- benchmarksof 12
- 12
- per 1,000 documentsmean over its benchmarks
- $1.12
- seconds per inputmean response time
- 17.5
By task
-
Data correction
60.6
Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
-
Document understanding
53.8
Best of 107 models with all its benchmarks: gpt-6-astra, 97.0.
-
Information extraction
73.3
Best of 107 models with all its benchmarks: grok-4.6, 90.8.
-
Transcription
76.1
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 Dataother prompt | 91.0±6.6 | ≈74of 115 | $2.19$0.011 run | 69.12 | T1998 · 24 Oct 2026prompt_v2.txt |
| Blacklist Cards | 92.9±3.7 | ≈51of 114 | $0.57$0.019 run | 4.67 | T2005 · 24 Oct 2026prompt.txt |
| Book Advert XML Repairother prompt | 25.0±9.5 | ≈64of 119 | $0.61$0.031 run | 9.56 | T2007 · 24 Oct 2026prompt_v2.txt |
| Business Letters | 50.0±7.1 | ≈100of 114 | $0.96$0.055 run | 5.08 | T1999 · 24 Oct 2026prompt.txt |
| Company Listsother prompt | 68.9±15.0 | ≈84of 114 | $0.94$0.014 run | 5.13 | T2006 · 24 Oct 2026prompt_v2.txt |
| Duty Rosters | 51.1±28.2 | ≈87of 107 | $1.61$0.034 run | 28.03 | T2011 · 24 Oct 2026prompt.txt |
| Fraktur Adverts | 75.2±36.8 | ≈32of 114 | $1.25$0.006 run | 14.54 | T2002 · 24 Oct 2026prompt_optimized.txt |
| General Meeting Minutes | 79.8±12.3 | ≈25of 108 | $2.91$0.026 run | 51.16 | T2010 · 24 Oct 2026prompt.txt |
| Library Cardsother prompt | 79.5±1.8 | ≈95of 114 | $0.23$0.060 run | 2.99 | T2003 · 24 Oct 2026prompt_year-string.txt |
| Magazine Pagesother prompt | 14.2±9.2 | ≈66of 115 | $0.55$0.025 run | 2.61 | T2009 · 24 Oct 2026prompt_explicit-json.txt |
| Medieval Manuscripts | 56.9±19.4 | ≈72of 114 | $0.61$0.007 run | 9.43 | T2004 · 24 Oct 2026prompt.txt |
| Personnel Cards | 96.2±1.3 | ≈30of 114 | $1.00$0.061 run | 8.27 | T2008 · 24 Oct 2026prompt.txt |
Price
| From | Input, $ per million tokens | Output, $ per million tokens | Source |
|---|---|---|---|
| 24 Oct 2026 | 0.15 | 0.6 | archived price page |
Every run
- T2011Duty Rosters24 Oct 202651.1$0.034
- T2010General Meeting Minutes24 Oct 202679.8$0.026
- T2009Magazine Pages24 Oct 202614.2$0.025
- T2008Personnel Cards24 Oct 202696.2$0.061
- T2007Book Advert XML Repair24 Oct 202625.0$0.031
- T2006Company Lists24 Oct 202668.9$0.014
- T2005Blacklist Cards24 Oct 202692.9$0.019
- T2004Medieval Manuscripts24 Oct 202656.9$0.007
- T2003Library Cards24 Oct 202679.5$0.060
- T2002Fraktur Adverts24 Oct 202675.2$0.006
- T2001Business Letters24 Oct 2026 · 32 signed letters57.0$0.022
- T2000Business Letters24 Oct 2026 · 25 unsigned letters51.0$0.010
- T1999Business Letters24 Oct 2026 · all 57 letters50.0$0.055
- T1998Bibliographic Data24 Oct 202691.0$0.011