AI Chat Cost Optimization Local Inference Mistral Multimodal Benchmarks

Mistral Small 1.0

Language model release

Visit site

Official Mistral AI language-model release in the Mistral Small line.

Pricing

Model profile · Current snapshot

Input / 1M

$0.075

Output / 1M

$0.19

Blended / 1M

$0.104

Output TPS

153

TTFT (s)

0.55

Model Intelligence

Arena ranking

Benchmarkable

Yes

Model level

release

Intelligence Index

12.7

Math Index

4.3

MMLU Pro

0.65

GPQA

0.46

HLE

0.04

LiveCodeBench

0.25

SciCode

0.24

MATH-500

0.72

AIME

0.08

AIME 2025

0.04

IFBench

0.26

LCR

TAU2

0.2

Recent stories

1 linked story

releaseSECONDARY2026-03-20

Mistral releases Small 4 with 256K context, image input, and $0.15/$0.6 pricing

Mistral Small 4 combines reasoning and non-reasoning modes in one 119B MoE, adds native image input, and expands context to 256K at $0.15/$0.6 per million tokens. It improves sharply over Small 3.2, but still trails similarly sized open peers on several evals.