Skip to content

magistral-small-2509

Mistral

legacy model Its test configurations are retired, so it is no longer run. The figures are from its best legacy runs, scored as they ran at the time; it is not ranked.

average±5.1 (95%): 28.5
benchmarksof 12: 11
per 1,000 documentsmean over its benchmarks: $1.16
seconds per inputmean response time: 13.4

Compare with other models Its row in the Matrix All its test runs

By task

Data correction

17.6

Best of 114 models with all its benchmarks: claude-opus-5-5, 82.7.
Transcription

20.1

Best of 114 models with all its benchmarks: gemini-4-argon, 96.2.

By benchmark

Benchmark	Score	Rank	Per 1,000 docs	s / input	Best run
Bibliographic Data	94.4±2.8	–	$3.77$0.019 run	14.98	T0565 · 4 Jun 2026prompt.txt
Blacklist Cards	5.1±6.0	–	$0.15$0.005 run	7.85	T0572 · 4 Jun 2026prompt.txt
Book Advert XML Repair	7.4±4.8	–	$0.73$0.037 run	3.41	T0575 · 25 Jan 2026prompt.txt
Business Letters	50.0±11.7	–	$1.77$0.035 run	3.36	T0566 · 4 Jun 2026prompt.txt
Company Listsother prompt	9.3±13.4	–	$0.27$0.004 run	8.3	T0574 · 4 Jun 2026prompt_min.txt
Fraktur Adverts	11.8±23.1	–	$0.90$0.004 run	65.13	T0569 · 4 Jun 2026prompt_optimized.txt
General Meeting Minutes	79.9±9.6	–	$2.64$0.024 run	12.32	T1236 · 3 Jul 2026prompt.txt
Library Cards	4.7±2.2	–	$0.076$0.020 run	7.19	T0570 · 4 Jun 2026prompt.txt
Magazine Pages	2.2±4.3	–	$0.27$0.012 run	7.31	T0810 · 4 Jun 2026prompt.txt
Medieval Manuscripts	20.7±40.6	–	$1.74$0.007 run	8.2	T0571 · 4 Jun 2026prompt.txt
Personnel Cards	27.7±22.6	–	$0.44$0.027 run	8.99	T0596 · 4 Jun 2026prompt.txt

Not run on: Duty Rosters.

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
2 Mar 2026	0.5	1.5	archived price page
23 Jan 2026	0.5	0.5	archived price page

Every run