Skip to content

Mistral AI · Chat

Codestral 25.01

Mistral AI 🇫🇷ChatOne of its sources says the weights aren't publicly available.The maker currently plans to deprecate it within the next 180 days (15 Feb 2027). Plans can change.

Our cutting-edge language model for coding released August 2025.

llminfo IQ
116.6
Maker's price
$0.45 blended / 1M
Max context
256K
Listings
39

Evidence and sources

llminfo IQ 116.6, ranked among chat models. From 10 benchmarks. A fair amount of data behind it.How llminfo IQ is computedPoor value (37.8). Costs more per task than its benchmark results predict. From 4 benchmarks with measured task costs. A fair amount of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsTerse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.Median 2,049 output tokens per task at default effort.How the Yap score worksReleased 29 May 2024 Training-data cutoff, as its sources state it Maker HQ: FR (according to Mistral AI's own website)
All rankings: #374 on OpenRouter today #377 on OpenRouter this week #393 on OpenRouter this month #506 on NanoGPT recentlySources: Mistral API data, Models.dev data, GenAI Prices data, LiteLLM data, AI/ML API data, OpenRouter data docs, llminfo researchPrimary data from Mistral API from Models.dev: name, weights availability from GenAI Prices: variant from LiteLLM: deprecation date from AI/ML API: primary mode from OpenRouter: knowledge cutoff from llminfo research: release date
Tags from its sources:instructtoolscodefimfunction-calling

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

Deprecation scheduled

The maker plans to deprecate it on 15 Feb 2027, within the next 180 days.

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 8 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Text
Files
Text

Capabilities

Supported operations and provider-stated features

Function callingStructured outputPrompt cachingPDF input

Listings

 Mistral AIcodestral-2508 · 2508Sources: GenAI Prices data, LiteLLM data, Mistral API data, Models.dev data128K$0.30cache read $0.03$0.90—
 Googlevertex_ai/codestral-2501 · 2501Source: LiteLLM data128K$0.20$0.60—
 TokenmixcodestralSources: TokenMix data, TokenMix PricingNot published by the provider$0.279$0.837—
 Microsoft Azureazure_ai/Codestral-2501 · 2501Sources: LiteLLM data, Models.dev data256K$0.30$0.90—
 Eden AIInference by Mistral AImistral/codestral-2508 · 2508Source: Models.dev data256K$0.30cache read $0.03$0.90—
 Flowdotcodestral-2508 · 2508Source: FlowDot data256K$0.30$0.90—
 Kilomistralai/codestralSources: Kilo AI Gateway data, Models.dev data128K$0.30cache read $0.03$0.90$0.00
 Llmgatewaycodestral-2508 · 2508Source: LLMGateway data256K$0.30$0.90—
 NanoGPTmistralai/codestral-2508 · 2508Sources: Models.dev data, NanoGPT256K$0.30cache read $0.15$0.90—
 Nous Portalmistralai/codestral-2508 · 2508Source: Nous Portal Inference Gateway data256K$0.30cache read $0.03$0.90—
Show all 32 listings

Showing 10 of 32: the maker's own listing first, then the cheapest per 1M input tokens from each provider.

Also sold by Aihubmix, Aimlapi, Cortecs AI, Eurouter, Llmgateway Providers, Lowrouter, Merge Gateway, OpenRouter, Requesty, Vercel, Vercel AI Gateway.

7 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Coding

Reasoning

Other

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

Around the web

  • README.md · mistralai/Codestral-22B-v0.1 at main - Hugging Face
    result = tokenizer.decode(out_tokens[ 0 ]) print (result) Inference with hugging face transformers from transformers import …result = tokenizer.decode(out_tokens[ 0 ]) print (result) Inference with hugging face transformers from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained( "mistral…

    huggingface.co

  • mistralai/Codestral-22B-v0.1 · Hugging Face
    mistralai/Codestral-22B-v0.1 · Hugging Face Codestral-22B-v0.1 is trained on a diverse dataset of 80+ programming languages, …mistralai/Codestral-22B-v0.1 · Hugging Face Codestral-22B-v0.1 is trained on a diverse dataset of 80+ programming languages, including the most popular ones, such as Python, Java, C, C++, JavaScript,…

    huggingface.co

  • mistralai/Codestral-22B-v0.1 at main - Hugging Face
    mistralai / Codestral-22B-v0.1 like 1.34k Follow Mistral AI_ 18.8k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.34k Follow Mistral AI_ 18.8k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 Copy to bucket new m…

    huggingface.co

  • Commits · mistralai/Codestral-22B-v0.1
    mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.4k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.4k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 83e1c24 Codestral-22…

    huggingface.co

  • mistralai/Codestral-22B-v0.1 · License
    mistralai / Codestral-22B-v0.1 like 1.34k Follow Mistral AI_ 18.8k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.34k Follow Mistral AI_ 18.8k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 Copy to bucket new L…

    huggingface.co

  • mistralai/Codestral-22B-v0.1 · Disclosing training data - Hugging Face
    mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.9k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.9k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 Disclosing training …

    huggingface.co

  • mistralai/Codestral-22B-v0.1 · Align tokenizer with mistral-common
    mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.9k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.33k Follow Mistral AI_ 15.9k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 Align tokenizer with…

    huggingface.co

  • mistralai/codestral-22b-v0.1 • LM Studio
    ← All Models Star 36 codestral-22b-v0.1 Public Fork 0 Star 36 Use Model in LM …← All Models Star 36 codestral-22b-v0.1 Public Fork 0 Star 36 Use Model in LM Studio Model Revisions Description Mistral AI's latest coding model, Codestral can handle both instructions and code comp…

    lmstudio.ai

  • mistralai/Codestral-22B-v0.1 · Fill in the Middle evaluation benchmark
    mistralai / Codestral-22B-v0.1 like 1.32k Follow Mistral AI_ 14.2k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.32k Follow Mistral AI_ 14.2k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 Fill in the Middle e…

    huggingface.co

  • mistralai/Codestral-22B-v0.1 · What is the context size on this model? And it does not appear to deal with JSON, function calling well.
    mistralai / Codestral-22B-v0.1 like 1.32k Follow Mistral AI_ 14.4k Safetensors code vllm mistral mistral-common License: …mistralai / Codestral-22B-v0.1 like 1.32k Follow Mistral AI_ 14.4k Safetensors code vllm mistral mistral-common License: mnpl Model card Files Files and versions xet Community 56 What is the context …

    huggingface.co

Also known as: aihubmix-codestral-2501, azure_ai/codestral-2501, codestral-2501, codestral-2508, codestral-latest, codestral/codestral-2405, codestral/codestral-latest, mistral/codestral, mistral/codestral-2508, mistral/codestral-2508@eu, mistral/codestral-latest, mistral/codestral-latest@eu, mistralai/codestral, mistralai/codestral-2501, mistralai/codestral-2508, text-completion-codestral/codestral-2405, text-completion-codestral/codestral-latest, vercel_ai_gateway/mistral/codestral, vertex_ai/codestral-2501

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works