Mistral Medium

Last updated: Recently

Model Type

MText

Max Context

262K

Max Output

Parameter Scale

Mistral Medium 3.5 is dense 128B instruction following model from Mistral AI. 262,144 token context window. Includes independent benchmarks from Artificial Analysis.

Input Modality

text, image, file

Output Modality

text

Inference Speed

73.210 tokens/s

Latest Release Date

12/11/2023

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

Mistral Medium

Release date

12/11/2023

Performance

73.210 tokens/s

model_creator.name

Mistral

AA Evaluation Scores

AA Intelligence Index

AA Coding Index

AA Math Index

MMLU Pro

0.491

GPQA

0.349

HLE

0.034

LiveCodeBench

0.099

SciCode

0.118

Math-500

0.404666666666667

AIME

0.0366666666666667

AIME 2025

IFBench

LCR

TerminalBench Hard

TAU2

Output Speed (tokens/s)

73.21 tok/s

TTFT (s)

0.56s

First Answer Token (s)

0.56s

Mistral

Mistral Medium

$2.75/M$8.10/M

Mistral Medium 3.5 is dense 128B instruction following model from Mistral AI. 262,144 token context window. Includes independent benchmarks from Artificial Analysis.

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0