Mistral

Mistral Large 4 Preview

Oct 6, 2026ReasoningClosed weightsModels.dev

Compare

Intelligence

25.3

#170 of 762Rank 170 of 762 measured models

Output speed

111 t/s

#168 of 420Rank 168 of 420 measured models

Price/1M

$2.07

#511 of 715Rank 511 of 715 measured models

Context window

524K

#590 of 1281Rank 590 of 1281 measured models

Cost per task

$5.54

#196 of 214Rank 196 of 214 measured models

Mistral Large 4 Preview scores 25.3 on the Artificial Analysis Intelligence Index, ranking 170th of 762 measured models. The closest models on this index are Hy3 (25.3), K2 Horizon MoVA 36B A4B (25.3), and Grok 4.20 0309 (25.2). At $2.07 per million tokens (3:1 input/output blended price), it is cheaper than Grok 4.20 0309 ($3), more expensive than Hy3 ($0.241). With an output speed of 111 t/s, it is faster than Hy3 (82.2 t/s).