Codestral 25.08

Codestral 25.08 is Mistral's API-only code model, released at the end of July 2025 for code completion. It supports fill-in-the-middle (FIM) completion through a dedicated endpoint, function calling and a 128K context. Mistral sells it as a premier model and does not publish its weights or parameter count.

By Mistral AIReleased July 30, 2025Data checked October 10, 2026

CurrentClosedEfficient

Specifications

Context window: 125Kabout 96,000 words, or 190 pages
Max output: Not published
Input price: $0.30per 1M tokens
Output price: $0.90per 1M tokens
Cached input: Not published
Weights: ClosedProprietary
Parameters: Undisclosed
Input types: Text
Output types: Text
Reasoning mode: No
Tool calling: Yes
Released: Jul 30, 2025Current

Mistral API price (codestral-2508); the docs show $0.3 and $0.9 per 1M tokens without labels, mapped as input then output. Parameter count is not published.

What it costs in practice

Workload	Monthly cost at list price
Support chatbot30,000 conversations a month, 2,000 input and 400 output tokens each	$29
Document summaries2,000 documents a month, 12,000 input and 800 output tokens each	$8.64
Coding agent1,000 tasks a month, 60,000 input and 6,000 output tokens each	$23

Standard rates, no caching or batch discounts. Run your own numbers.

Strengths

Dedicated fill-in-the-middle endpoint for IDE completion
128K context for repository-level prompts
Function calling support

Trade-offs to weigh

Weights are not available for self-hosting
Parameter count and architecture are not disclosed

Where it fits

Mistral AI positions Codestral 25.08 for:

codinglow latencyenterprise

Where you can use it

Alternatives to consider

Model	Tier	Context	Input / output per 1M
Gemini 3.1 Flash-LiteGoogle	Efficient	1M	$0.25 / $1.50	Compare
SonarPerplexity	Efficient	—	$0.25 / $2.50	Compare
Gemini 3.5 Flash-LiteGoogle	Efficient	1M	$0.30 / $2.50	Compare
Muse Glimmer 30BMeta	Efficient	128K	$0.35 / $1.50	Compare

Frequently asked questions

How much does Codestral 25.08 cost?

Codestral 25.08 lists at $0.30 per million input tokens and $0.90 per million output tokens (checked October 10, 2026). Mistral API price (codestral-2508); the docs show $0.3 and $0.9 per 1M tokens without labels, mapped as input then output. Parameter count is not published. Confirm current rates on the provider's pricing page before you commit.

What is the context window of Codestral 25.08?

Codestral 25.08 accepts up to 128,000 tokens of context (about 96,000 words, or 190 pages). Long prompts cost more and can be slower, so most apps send only what the task needs.

Is Codestral 25.08 open source?

No. Codestral 25.08 is a proprietary model from Mistral AI; its weights aren't published. You can use it through the Mistral AI API and through the provider's apps.

What is Codestral 25.08 best for?

Mistral AI positions Codestral 25.08 for coding, low latency, enterprise. It accepts text as input, doesn't offer a separate reasoning mode and supports tool calling. Test it on a sample of your own tasks before you choose it for production.

Sources

Facts checked against these pages on October 10, 2026. Spot an error? Tell us.

Still choosing a model?

Send us your use case and constraints. We'll come back with a shortlist and an evaluation plan you can run on your own data.

Free and vendor-neutral: no model provider pays for placement.
Your details go only to partners you agree to.