Devstral Small (Jul '25)

Last updated: Recently

Model Type

MText

Max Context

131K

Max Output

Parameter Scale

Devstral Small 1.1 is 24B parameter open weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI. 131,072 token context window. Includes independent benchmarks from Artificial Analysis.

Input Modality

text, file

Output Modality

text

Inference Speed

192.713 tokens/s

Latest Release Date

7/10/2025

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

Devstral Small (Jul '25)

Release date

7/10/2025

Performance

192.713 tokens/s

model_creator.name

Mistral

AA Evaluation Scores

AA Intelligence Index

15.2

AA Coding Index

12.1

AA Math Index

29.3

MMLU Pro

0.622

GPQA

0.414

HLE

0.037

LiveCodeBench

0.254

SciCode

0.243

Math-500

0.634666666666667

AIME

0.00333333333333333

AIME 2025

0.293333333333333

IFBench

0.345578231292517

LCR

0.17

TerminalBench Hard

0.0606060606060606

TAU2

0.283625730994152

Output Speed (tokens/s)

192.713 tok/s

TTFT (s)

0.402s

First Answer Token (s)

0.402s

Mistral

Devstral Small (Jul '25)

$0.10/M$0.30/M

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0