Mistral AI

Mistral Small 3.2 24B Instruct

Multimodal
Zero-eval
#1HumanEval Plus
#1IF
#1MBPP Plus
+1 more

by Mistral AI

+
+
+
+
About

Mistral Small 3.2 24B Instruct represents a further evolution of the Small model series, developed with continued refinements to instruction-following and task performance. Built to incorporate ongoing improvements, it provides the latest capabilities in Mistral's intermediate-scale offering.

+
+
+
+
Timeline
AnnouncedJun 20, 2025
ReleasedJun 20, 2025
Knowledge CutoffOct 1, 2023
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Apache 2.0
Base ModelMistral Small 3.1 24B Base
Performance Overview
Performance metrics and category breakdown

Overall Performance

15 benchmarks
Average Score
69.6%
Best Score
94.9%
High Performers (80%+)
6
+
+
+
+
All Benchmark Results for Mistral Small 3.2 24B Instruct
Complete list of benchmark scores with detailed information
DocVQA
multimodal
0.95
94.9%
Self-reported
AI2D
multimodal
0.93
92.9%
Self-reported
HumanEval Plus
text
0.93
92.9%
Self-reported
ChartQA
multimodal
0.87
87.4%
Self-reported
IF
text
0.85
84.8%
Self-reported
MMLU
text
0.81
80.5%
Self-reported
MBPP Plus
text
0.78
78.3%
Self-reported
MATH
text
0.69
69.4%
Self-reported
MMLU-Pro
text
0.69
69.1%
Self-reported
MathVista
multimodal
0.67
67.1%
Self-reported
Showing 1 to 10 of 15 benchmarks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+