Codestral 2508

mistralai/codestral-2508

Mistral: Codestral 2508, released by MistralAI on 2025‑08‑01, accepts text and file inputs and generates text output. It provides a context window of up to 256 000 tokens, offers open weights, supports tool usage but not reasoning, and is priced at $0.30 per input token and $0.90 per output token.

  • tool use
  • structured outputs

At a glance

providermistralai
releasedJul 30, 2025
context256K
input / M$0.30

Specifications

What the API gives you

Context window
256K tokens
Max output
205K tokens
Input price
$0.30 / 1M tokens
Output price
$0.90 / 1M tokens
Cache read
$0.030 / 1M tokens
Modalities
text, file → text
Knowledge cutoff
Mar 31, 2025
Released
Jul 30, 2025

Measured evidence

Public benchmark record

Design Arena Coding
864 Elo2026-07-28
Design Arena Coding
864 Elo2026-07-27
Design Arena Coding
864 Elo2026-07-30
Design Arena Coding
864 Elo2026-07-29
Design Arena Coding
864 Elo2026-07-31
Design Arena Coding
863 Elo2026-08-02
Design Arena Coding
862 Elo2026-08-05
Design Arena Coding
863 Elo2026-08-01
Design Arena Coding
861 Elo2026-08-09
Design Arena Coding
863 Elo2026-08-03
Design Arena Coding
863 Elo2026-08-04
Design Arena Coding
859 Elo2026-08-14
Design Arena Coding
863 Elo2026-08-06
Design Arena Coding
861 Elo2026-08-10
Design Arena Coding
863 Elo2026-08-07
Design Arena Coding
853 Elo2026-08-18
Design Arena Coding
862 Elo2026-08-08
Design Arena Coding
853 Elo2026-08-15
Design Arena Coding
861 Elo2026-08-11
Design Arena Coding
861 Elo2026-08-12
Design Arena Coding
859 Elo2026-08-13
Design Arena Coding
851 Elo2026-08-19
Design Arena Coding
858 Elo2026-08-27
Design Arena Coding
851 Elo2026-08-20
Design Arena Coding
853 Elo2026-08-16
Design Arena Coding
850 Elo2026-09-05
Design Arena Coding
852 Elo2026-08-17
Design Arena Coding
858 Elo2026-08-21
Design Arena Coding
858 Elo2026-08-22
Design Arena Coding
858 Elo2026-08-23
Design Arena Coding
858 Elo2026-08-28
Design Arena Coding
858 Elo2026-08-24
Design Arena Coding
859 Elo2026-08-25
Design Arena Coding
859 Elo2026-08-26
Design Arena Coding
850 Elo2026-09-06
Design Arena Coding
858 Elo2026-08-29
Design Arena Coding
857 Elo2026-08-30
Design Arena Coding
857 Elo2026-08-31
Design Arena Coding
857 Elo2026-09-01
Design Arena Coding
856 Elo2026-09-02
Design Arena Coding
854 Elo2026-09-03
Design Arena Coding
853 Elo2026-09-04
Design Arena UI
889 Elo2026-07-28
Design Arena UI
889 Elo2026-07-27
Design Arena UI
889 Elo2026-07-30
Design Arena UI
889 Elo2026-07-29
Design Arena UI
889 Elo2026-07-31
Design Arena UI
889 Elo2026-08-02
Design Arena UI
889 Elo2026-08-01
Design Arena UI
888 Elo2026-08-05
Design Arena UI
889 Elo2026-08-03
Design Arena UI
885 Elo2026-08-09
Design Arena UI
888 Elo2026-08-04
Design Arena UI
883 Elo2026-08-14
Design Arena UI
887 Elo2026-08-06
Design Arena UI
887 Elo2026-08-07
Design Arena UI
885 Elo2026-08-10
Design Arena UI
885 Elo2026-08-08
Design Arena UI
879 Elo2026-08-18
Design Arena UI
885 Elo2026-08-11
Design Arena UI
880 Elo2026-08-15
Design Arena UI
886 Elo2026-08-12
Design Arena UI
886 Elo2026-08-13
Design Arena UI
879 Elo2026-08-19
Design Arena UI
880 Elo2026-08-16
Design Arena UI
879 Elo2026-08-20
Design Arena UI
878 Elo2026-08-27
Design Arena UI
879 Elo2026-08-17
Design Arena UI
874 Elo2026-09-05
Design Arena UI
878 Elo2026-08-21
Design Arena UI
879 Elo2026-08-22
Design Arena UI
879 Elo2026-08-23
Design Arena UI
879 Elo2026-08-24
Design Arena UI
878 Elo2026-08-28
Design Arena UI
879 Elo2026-08-25
Design Arena UI
879 Elo2026-08-26
Design Arena UI
875 Elo2026-09-06
Design Arena UI
878 Elo2026-08-29
Design Arena UI
878 Elo2026-08-30
Design Arena UI
878 Elo2026-08-31
Design Arena UI
878 Elo2026-09-01
Design Arena UI
877 Elo2026-09-02
Design Arena UI
875 Elo2026-09-03
Design Arena UI
874 Elo2026-09-04
Design Arena Website
834 Elo2026-07-28
Design Arena Website
834 Elo2026-07-30
Design Arena Website
834 Elo2026-07-27
Design Arena Website
834 Elo2026-07-29
Design Arena Website
834 Elo2026-07-31
Design Arena Website
833 Elo2026-08-02
Design Arena Website
833 Elo2026-08-01
Design Arena Website
833 Elo2026-08-05
Design Arena Website
833 Elo2026-08-03
Design Arena Website
832 Elo2026-08-09
Design Arena Website
833 Elo2026-08-04
Design Arena Website
829 Elo2026-08-14
Design Arena Website
833 Elo2026-08-06
Design Arena Website
833 Elo2026-08-07
Design Arena Website
831 Elo2026-08-10
Design Arena Website
832 Elo2026-08-08
Design Arena Website
820 Elo2026-08-18
Design Arena Website
832 Elo2026-08-11
Design Arena Website
820 Elo2026-08-15
Design Arena Website
831 Elo2026-08-12
Design Arena Website
829 Elo2026-08-13
Design Arena Website
820 Elo2026-08-19
Design Arena Website
820 Elo2026-08-16
Design Arena Website
820 Elo2026-08-20
Design Arena Website
829 Elo2026-08-27
Design Arena Website
820 Elo2026-08-17
Design Arena Website
825 Elo2026-09-05
Design Arena Website
829 Elo2026-08-21
Design Arena Website
829 Elo2026-08-22
Design Arena Website
829 Elo2026-08-23
Design Arena Website
829 Elo2026-08-24
Design Arena Website
829 Elo2026-08-28
Design Arena Website
829 Elo2026-08-25
Design Arena Website
830 Elo2026-08-26
Design Arena Website
822 Elo2026-09-06
Design Arena Website
829 Elo2026-08-29
Design Arena Website
828 Elo2026-08-30
Design Arena Website
828 Elo2026-08-31
Design Arena Website
829 Elo2026-09-01
Design Arena Website
827 Elo2026-09-02
Design Arena Website
826 Elo2026-09-03
Design Arena Website
825 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.

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.