Skip to content

mistral-medium-2508

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: 58.4
benchmarksof 12: 11
per 1,000 documentsmean over its benchmarks: $2.02
seconds per inputmean response time: 77.0

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

By task

Data correction

35.5

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

55.8

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	95.0±4.2	–	$4.91$0.025 run	185.61	T0169 · 4 Jun 2026prompt.txt
Blacklist Cardsolder run	89.8	–	$0.93$0.031 run	5.17	T0327 · 24 Oct 2025prompt.txt
Book Advert XML Repair	6.4±3.5	–	$1.66$0.083 run	6.12	T0470 · 24 Jan 2026prompt.txt
Business Lettersolder run	48.0	–	–– run	3.07	T0171 · 26 Sep 2025prompt.txt
Company Lists	71.7±14.1	–	$2.88$0.043 run	63.46	T0383 · 4 Jun 2026prompt.txt
Fraktur Advertsolder run	35.3	–	–– run	31.78	T0177 · 26 Sep 2025prompt_optimized.txt
General Meeting Minutes	76.8±10.9	–	$2.88$0.026 run	19.24	T1241 · 3 Jul 2026prompt.txt
Library Cardsolder run	78.2	–	$1.02$0.27 run	3.55	T0179 · 1 Oct 2025prompt.txt
Magazine Pages	9.8±8.3	–	$0.58$0.026 run	199.27	T0781 · 4 Jun 2026prompt.txt
Medieval Manuscripts	67.3±15.1	–	$1.60$0.018 run	144.5	T0295 · 4 Jun 2026prompt.txt
Personnel Cards	64.7±12.4	–	$1.70$0.10 run	185.1	T0602 · 4 Jun 2026prompt.txt

Not run on: Duty Rosters.

Price

From	Input, $ per million tokens	Output, $ per million tokens	Source
0 Sep 2025	0.4	2.0	archived price page

Every run