Mistral Small 3.2
MistralMistralOpen WeightApache 2.0 · Commercial OK
Description
Mistral-Small-3.2-24B-Instruct-2506 is a minor update of Mistral-Small-3.1-24B-Instruct-2503.
Release Date
2025-06-20
Parameters
23.6B
Context Length
128K
Modalities
image, text
Capability Radar
29
general
20
coding
40
reasoning
33
scienceest.
0
agents
90
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 324 | 22.0 | AA |
| General Ranking | 327 | 34.0 | AA |
| Math Reasoning | 218 | 41.0 | AA |
| Multimodal Ranking | 5 | 92.0 | LS |
| Science | 331 | 33.0 | AA |
Benchmark Scores (LLM Stats)
Biology
GPQA
46.1%SR
Code
HumanEval Plus
92.9%SR
MBPP Plus
78.3%SR
Communication
Wild Bench
65.3%SR
Creativity
Arena Hard
43.1%SR
Factuality
SimpleQA
12.1%SR
Finance
MMLU
80.5%SR
MMLU-Pro
69.1%SR
General
IF
84.8%SR
MMMU
62.5%SR
Image To Text
DocVQA
94.9%SR
Math
MATH
69.4%SR
MathVista
67.1%SR
Multimodal
AI2D
92.9%SR
ChartQA
87.4%SR
AA Evaluation Indices
Math Index27.0
Intelligence Index15.1
Coding Index13.3
Math 5000.9
Mmlu Pro0.7
Gpqa0.5
Ifbench0.3
Aime0.3
Tau20.3
Livecodebench0.3
Aime 250.3
Scicode0.3
Lcr0.2
Terminalbench Hard0.1
Hle0.0
LLM Stats Category Scores
Image To Text90
Vision80
Multimodal80
Communication70
Finance70
Healthcare70
Language70
Legal70
Math70
Reasoning60
Biology50
Chemistry50
General50
Physics50
Writing40
Creativity40
Factuality10
Pricing
Input Price$0.087 / 1M tokens
Output Price$0.25 / 1M tokens
Blended Price (3:1)$0.128 / 1M tokens
Speed
Tokens/sec143.8 tokens/s
Time to First Token0.38s
Time to Answer0.38s
Available Providers
(LS internal units)No provider data available