Mistral Small 3.2 24B

mistralai/mistral-small-3.2-24b-instruct-2506

Mistral Small 3.2 24B, released by Mistral AI on 2025‑06‑20, accepts both image and text inputs and generates text outputs. It supports tool usage, does not support reasoning, and provides an open‑weights release with a context window of 131,072 tokens. The model is priced at 0.09999999999999999 per input token and 0.3 per output token.

  • tool use
  • structured outputs

At a glance

providermistralai
releasedJun 20, 2025
context131K
input / M$0.075

Specifications

What the API gives you

Context window
131K tokens
Max output
16K tokens
Input price
$0.075 / 1M tokens
Output price
$0.20 / 1M tokens
Cache read
Modalities
image, text → text
Knowledge cutoff
Oct 31, 2023
Released
Jun 20, 2025

Measured evidence

Public benchmark record

Design Arena Coding
937 Elo2026-07-28
Design Arena Coding
938 Elo2026-07-27
Design Arena Coding
937 Elo2026-07-30
Design Arena Coding
937 Elo2026-07-29
Design Arena Coding
937 Elo2026-07-31
Design Arena Coding
937 Elo2026-08-02
Design Arena Coding
936 Elo2026-08-05
Design Arena Coding
937 Elo2026-08-01
Design Arena Coding
935 Elo2026-08-09
Design Arena Coding
937 Elo2026-08-03
Design Arena Coding
936 Elo2026-08-04
Design Arena Coding
932 Elo2026-08-14
Design Arena Coding
936 Elo2026-08-06
Design Arena Coding
935 Elo2026-08-10
Design Arena Coding
936 Elo2026-08-07
Design Arena Coding
926 Elo2026-08-18
Design Arena Coding
935 Elo2026-08-08
Design Arena Coding
926 Elo2026-08-15
Design Arena Coding
935 Elo2026-08-11
Design Arena Coding
934 Elo2026-08-12
Design Arena Coding
932 Elo2026-08-13
Design Arena Coding
925 Elo2026-08-19
Design Arena Coding
931 Elo2026-08-27
Design Arena Coding
925 Elo2026-08-20
Design Arena Coding
926 Elo2026-08-16
Design Arena Coding
923 Elo2026-09-05
Design Arena Coding
926 Elo2026-08-17
Design Arena Coding
931 Elo2026-08-21
Design Arena Coding
932 Elo2026-08-22
Design Arena Coding
932 Elo2026-08-23
Design Arena Coding
931 Elo2026-08-28
Design Arena Coding
932 Elo2026-08-24
Design Arena Coding
932 Elo2026-08-25
Design Arena Coding
932 Elo2026-08-26
Design Arena Coding
923 Elo2026-09-06
Design Arena Coding
931 Elo2026-08-29
Design Arena Coding
931 Elo2026-08-30
Design Arena Coding
931 Elo2026-08-31
Design Arena Coding
931 Elo2026-09-01
Design Arena Coding
929 Elo2026-09-02
Design Arena Coding
927 Elo2026-09-03
Design Arena Coding
926 Elo2026-09-04
Design Arena UI
943 Elo2026-07-28
Design Arena UI
944 Elo2026-07-27
Design Arena UI
944 Elo2026-07-30
Design Arena UI
944 Elo2026-07-29
Design Arena UI
943 Elo2026-07-31
Design Arena UI
944 Elo2026-08-02
Design Arena UI
944 Elo2026-08-01
Design Arena UI
943 Elo2026-08-05
Design Arena UI
944 Elo2026-08-03
Design Arena UI
940 Elo2026-08-09
Design Arena UI
943 Elo2026-08-04
Design Arena UI
938 Elo2026-08-14
Design Arena UI
942 Elo2026-08-06
Design Arena UI
942 Elo2026-08-07
Design Arena UI
940 Elo2026-08-10
Design Arena UI
940 Elo2026-08-08
Design Arena UI
934 Elo2026-08-18
Design Arena UI
940 Elo2026-08-11
Design Arena UI
935 Elo2026-08-15
Design Arena UI
940 Elo2026-08-12
Design Arena UI
941 Elo2026-08-13
Design Arena UI
934 Elo2026-08-19
Design Arena UI
935 Elo2026-08-16
Design Arena UI
934 Elo2026-08-20
Design Arena UI
933 Elo2026-08-27
Design Arena UI
934 Elo2026-08-17
Design Arena UI
929 Elo2026-09-05
Design Arena UI
933 Elo2026-08-21
Design Arena UI
934 Elo2026-08-22
Design Arena UI
934 Elo2026-08-23
Design Arena UI
934 Elo2026-08-24
Design Arena UI
933 Elo2026-08-28
Design Arena UI
934 Elo2026-08-25
Design Arena UI
934 Elo2026-08-26
Design Arena UI
930 Elo2026-09-06
Design Arena UI
933 Elo2026-08-29
Design Arena UI
933 Elo2026-08-30
Design Arena UI
933 Elo2026-08-31
Design Arena UI
933 Elo2026-09-01
Design Arena UI
932 Elo2026-09-02
Design Arena UI
930 Elo2026-09-03
Design Arena UI
929 Elo2026-09-04
Design Arena Website
920 Elo2026-07-28
Design Arena Website
920 Elo2026-07-30
Design Arena Website
920 Elo2026-07-27
Design Arena Website
920 Elo2026-07-29
Design Arena Website
920 Elo2026-07-31
Design Arena Website
919 Elo2026-08-02
Design Arena Website
919 Elo2026-08-01
Design Arena Website
919 Elo2026-08-05
Design Arena Website
919 Elo2026-08-03
Design Arena Website
918 Elo2026-08-09
Design Arena Website
919 Elo2026-08-04
Design Arena Website
915 Elo2026-08-14
Design Arena Website
919 Elo2026-08-06
Design Arena Website
919 Elo2026-08-07
Design Arena Website
918 Elo2026-08-10
Design Arena Website
918 Elo2026-08-08
Design Arena Website
907 Elo2026-08-18
Design Arena Website
918 Elo2026-08-11
Design Arena Website
907 Elo2026-08-15
Design Arena Website
917 Elo2026-08-12
Design Arena Website
916 Elo2026-08-13
Design Arena Website
907 Elo2026-08-19
Design Arena Website
906 Elo2026-08-16
Design Arena Website
907 Elo2026-08-20
Design Arena Website
915 Elo2026-08-27
Design Arena Website
907 Elo2026-08-17
Design Arena Website
911 Elo2026-09-05
Design Arena Website
915 Elo2026-08-21
Design Arena Website
915 Elo2026-08-22
Design Arena Website
916 Elo2026-08-23
Design Arena Website
916 Elo2026-08-24
Design Arena Website
915 Elo2026-08-28
Design Arena Website
916 Elo2026-08-25
Design Arena Website
916 Elo2026-08-26
Design Arena Website
909 Elo2026-09-06
Design Arena Website
915 Elo2026-08-29
Design Arena Website
915 Elo2026-08-30
Design Arena Website
915 Elo2026-08-31
Design Arena Website
915 Elo2026-09-01
Design Arena Website
914 Elo2026-09-02
Design Arena Website
912 Elo2026-09-03
Design Arena Website
911 Elo2026-09-04

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference.

Open-weight model on Hugging Face.

Task evidence

How broad is the support?

Each status is exact to this model version and the selected task. Official release claims remain provisional.

  • Best LLM for Frontend & UIProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Coding or LMArena WebDev. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for CodingProvisional evidence

    24% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 8 of 8 configured evaluation feeds currently have results.

  • Best Cheap LLM for High VolumeProvisional evidence

    17% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Cheapest Capable LLMProvisional evidence

    17% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Fast SummariesProvisional evidence

    14% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Best Value LLMProvisional evidence

    14% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • 13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Route reliability. 2 of 4 configured evaluation feeds currently have results.

  • Best LLM for Code CompletionProvisional evidence

    13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Best LLM for TranslationProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 6 of 6 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 5 configured evaluation feeds currently have results.

  • Best LLM for Business WritingProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Creative Writing. 5 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Marketing CopyProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Creative Writing. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for SummarizationProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • Best LLM for Chat & RoleplayProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Creative WritingProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for OCR & DocumentsProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Reasoning or LMArena Vision. 3 of 4 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for WritingProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Grounded RAGProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Charts & DiagramsProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Document ParsingProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • 8% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Massive ContextProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Long Query. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Hermes AgentProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for OpenHandsProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for GooseProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • 6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.

  • Best LLM for OpenClawProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • 6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.

  • 6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.

  • Best Local LLM for CodingProvisional evidence

    6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Competition MathProvisional evidence

    5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Agentic CodingProvisional evidence

    5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Design Arena Full-stack or Design Arena Web-apps. 8 of 8 configured evaluation feeds currently have results.

  • Best LLM for Autonomous AgentsProvisional evidence

    5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Frontier-Bench or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Math & ReasoningProvisional evidence

    5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for AI AgentsProvisional evidence

    5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.