Venice Atlas
← Back to model catalog

VENICE ATLAS · MODEL DOSSIER · MISTRAL SMALL 3.2 24B INSTRUCT

Model overview

Mistral Small 3.2 24B Instruct

Mistral Small 3.2 is a 24B parameter model optimized for efficiency and performance. Ideal for general-purpose tasks with balanced speed and capability.

mistral-small-3-2-24b-instruct · venice.ai
Text · Venice catalog$0.09 per 1M tokensPrice on Venice AI
ModalityTextOn Venice AI
Primary price$0.09 per 1M tokensper 1M tokens
Context window256KMax tokens
Benchmarks15Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Factuality#84.90 ratingImage To Text#820.19 rating
Factuality1 result
SimpleQAsimpleqa · factuality
12.1%
ReportedOct 2026
General intelligence4 results
Arena Hardarena-hard · general
43.1%
ReportedOct 2026
IFif · general
84.8%
ReportedOct 2026
MMLUmmlu · general
80.5%
ReportedOct 2026
Wild Benchwild-bench · general
65.3%
ReportedOct 2026
Language1 result
MMLU-Prommlu-pro · language
69.1%
ReportedOct 2026
Math2 results
MATHmath · math
69.4%
ReportedOct 2026
MathVistamathvista · math
67.1%
ReportedOct 2026
Multimodal1 result
MMMUmmmu · multimodal
62.5%
ReportedOct 2026
Reasoning4 results
ChartQAchartqa · reasoning
87.4%
ReportedOct 2026
GPQAgpqa · reasoning
46.1%
ReportedOct 2026
HumanEval Plushumaneval-plus · reasoning
92.9%
ReportedOct 2026
MBPP Plusmbpp-plus · reasoning
78.3%
ReportedOct 2026
Vision2 results
AI2Dai2d · vision
92.9%
ReportedOct 2026
DocVQAdocvqa · vision
94.9%
ReportedOct 2026

Learn what each benchmark measures ↗

Related models

More from mistral

mistral-small-2603Mistral Small 4$0.19 input per 1M tokens ↗