Mistral Small 3.1 24B (2503)

Building upon Mistral Small 3 (2501), Mistral Small 3.1 (2503) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance. With 24 billion parameters, this model achieves top-tier capabilities in both text and vision tasks.

Added Apr 15, 2025

Pricing

Auto routing · per 1M tokens

Input: $0.10
Output: $0.30

Compare provider prices

Specifications

Context window: 128K
Max output: 131.1K
Parameters: 24B
Avg output (7d): 109 tokens; Longer than 6% of models

Benchmarks

Artificial Analysis

LMArena

Sourced from Artificial Analysis.

Intelligence Index

7.1

Better than 22% of models compared

Coding Index

26.3

Better than 33% of models compared

Agentic work

AutomationBench-AA

Workflow automation with guardrail penalties

1.2%

Better than 16% of models compared

AA-Briefcase

Agentic knowledge work (Elo)

313 Elo

Better than 17% of models compared

GDPval-AA v2

Economically valuable tasks (Elo)

363 Elo

Better than 21% of models compared

Document reasoning

GDP.pdf

Professional PDF reasoning: all-pass rate

1.0%

Better than 7% of models compared

AA-LCR v1.1

Long context reasoning with updated grading

22.3%

Better than 24% of models compared

Reasoning

HLE

Humanity's Last Exam

4.3%

Better than 18% of models compared

IFBench

Instruction-following benchmark

29.9%

Better than 14% of models compared

Coding

Terminal-Bench v4.0

Practical coding and terminal tasks

0.0%

Better than 16% of models compared

SciCode

Python programming for scientific computing

27.8%

Better than 8% of models compared

LiveCodeBench

Contamination-free coding benchmark

21.2%

Better than 23% of models compared

Math

AIME 2025

American Invitational Mathematics Examination 2025

3.7%

Better than 6% of models compared

AIME

American Invitational Mathematics Examination

9.3%

Better than 29% of models compared

Math-500

Diverse mathematical problem solving benchmark

70.7%

Better than 30% of models compared

Knowledge

MMLU-Pro

Professional and academic subject knowledge

65.9%

Better than 28% of models compared

Legacy benchmarks

GPQA Diamond (legacy)

Graduate-level scientific reasoning

45.4%

Better than 20% of models compared

Terminal-Bench Hard (legacy)

Agentic coding and terminal use

7.6%

Better than 40% of models compared

T²-Bench Telecom (legacy)

Conversational AI agents in dual-control scenarios

25.1%

Better than 29% of models compared

AA-LCR (unversioned / legacy)

Long context reasoning evaluation