Mistral

Mistral Medium 3

Mistral Medium 3 is a balanced, mid-sized model from Mistral AI, designed to offer strong performance across reasoning, coding, and multimodal tasks while maintaining efficiency. It aims to provide a good trade-off between capability, speed, and cost for a wide range of applications.

CodingReasoningFastCheapMultimodal

Input / 1M tokens

$0.40

Output / 1M tokens

$2.00

Output tokens/s

37.2

First-token seconds

0.55s

Supported plans

Benchmark history

Evaluations

TAU2

Measured May 29, 2026Source

Score

0.24

Terminalbench Hard

Measured May 29, 2026Source

Score

0.04

Lcr

Measured May 29, 2026Source

Score

0.28

Ifbench

Measured May 29, 2026Source

Score

0.39

Aime 25

Measured May 29, 2026Source

Score

0.3

Aime

Measured May 29, 2026Source

Score

0.44

Math 500

Measured May 29, 2026Source

Score

0.91

Scicode

Measured May 29, 2026Source

Score

0.33

Livecodebench

Measured May 29, 2026Source

Score

0.4

Hle

Measured May 29, 2026Source

Score

0.04

Gpqa

Measured May 29, 2026Source

Score

0.58

Mmlu Pro

Measured May 29, 2026Source

Score

0.76

Artificial Analysis Math Index

Measured May 29, 2026Source

Score

30.3

Artificial Analysis Coding Index

Measured May 29, 2026Source

Score

13.6

Artificial Analysis Intelligence Index

Measured May 29, 2026Source

Score

18.8

Plan availability

Products and plans that support this model

No products or plans have been linked to this model yet.

User ratings

Loading ratings...

Discussion

Thinking... Make sure you are connected to GitHub server