Mistral Small 3.1

Creator

Mistral AI

Released

2025-03-17

Intelligence

14.7

Artificial Analysis Index

Coding

26.3

Artificial Analysis Index

In $/1M

$0.10

input tokens

Out $/1M

$0.30

output tokens

Blended $/1M

$0.15

3:1 blended

Speed

0

tokens / sec

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
45
Humanity’s Last Exam
5
MMLU-Pro
66
SciCode
27
LiveCodeBench
21
MATH-500
71
AIME 2025
4
τ²-Bench (agentic)
25
Terminal-Bench Hard
8
IFBench
30
Via Artificial Analysis
© 2026 NYSGPT2525 LLC