Models Model details

Mistral AI

Mistral: Mistral Small 3.2 24B

mistralai/mistral-small-3.2-24b-instruct

Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on WildBench and Arena Hard, reduces infinite generations, and delivers gains in tool use and structured output tasks. It supports image and text inputs with structured outputs, function/tool calling, and strong performance across coding (HumanEval+, MBPP), STEM (MMLU, MATH, GPQA), and vision benchmarks (ChartQA, DocVQA).

Input
$0.09375

/ 1M tokens

Output
$0.25

/ 1M tokens

Context window
256K

tokens

Limits

Context

256,000

tokens

Max output

16,384

6.4% of the context window

Parameters

24,011,361,280

Model facts

Author

Mistral AI

Tokenizer

Mistral

Modality

Text + Image -> Text

License

Apache-2.0

Knowledge cutoff

2023-10-31

Created

Jun 20, 2025

Hugging Face

mistralai/Mistral-Small-3.2-24B-Instruct-2506

Default generation parameters

temperature

0.3

Publication status

Hugging Face: verifiedLicense: publishedParameter count: published

Modalities

Input · ImageInput · TextOutput · Text

Generation controls

Frequency penaltyLogit biasLogprobsMax tokensMin pPresence penaltyRepetition penaltyResponse formatSeedStopStructured outputsTemperatureTool choiceToolsTop kTop logprobsTop p

Creative tasks

Category performance

Data visualization
Top 97%

929 Elo

43.3% win

UI components
Top 98%

906 Elo

40.5% win

Code
Top 98%

909 Elo

39.8% win

Games
Top 99%

890 Elo

39.4% win

Websites
Top 96%

897 Elo

38.3% win