Mistral

Mistral Small 3.2

Jun 20, 2025Open weightsModels.devHugging Face

Compare

Intelligence

7.0

#571 of 744Rank 571 of 744 measured models

Coding

12.5

#258 of 298Rank 258 of 298 measured models

Output speed

37 t/s

#375 of 403Rank 375 of 403 measured models

Price/1M

$0.106

#96 of 659Rank 96 of 659 measured models

Context window

128K

#773 of 1199Rank 773 of 1199 measured models

Cost per task

$0.109

#54 of 193Rank 54 of 193 measured models

Mistral Small 3.2 scores 7.0 on the Artificial Analysis Intelligence Index, ranking 571st of 744 measured models. The closest models on this index are Command A+ (7.0), GPT-4 Turbo (7.0), and Nova Pro (7.0). At $0.106 per million tokens (3:1 input/output blended price), it is cheaper than Command A+ ($4.38), GPT-4 Turbo ($15), and Nova Pro ($1.40). With an output speed of 37 t/s, it is faster than GPT-4 Turbo (32.3 t/s), slower than Command A+ (61.5 t/s) and Nova Pro (127 t/s).