All Mistral AI models
M

Codestral 2508

Mistral AI · released August 2025

Codestral 2508 is a Mistral AI model with a 256K token context window — about 370 pages of text. It reads Text and Files and replies in Text.

$0.345

input / 1M tokens

$1.03

output / 1M tokens

256K

context — about 370 pages

0.26¢

a client proposal

Pricing

Input
$0.345 per million tokens
Output
$1.03 per million tokens
Cached input
$0.035 per million tokens

Billed from your Alyph wallet. Prices can change when the provider changes theirs — last checked August 9, 2026.

The numbers

Provider
Mistral AI
Released
August 2025 (2025-08-01)
Context window
256K tokens — about 370 pages
Max output
Not specified
Input
Text and Files
Output
Text
Reasoning
Answers directly
Knowledge cutoff
2025-03-31

Cost vs Token Amount

Drag the slider to adjust the split between input and output workload.

80% Input20% Output
free$0.24$0.480 tokens250k500k750k1000k
Codestral 2508

Adjust the ratio slider to change how output tokens influence the final price for equivalent workloads.

What a task costs

2K tokens in · 500 tokens out.

Codestral 2508

Mistral AI

0.12¢

for this task

57% input · 43% output

Full pricing →

Your $5 welcome credit covers about 4,140 of these.

Compare Codestral 2508

Closest in price on a typical task:

Codestral 2508 vs GLM 4.6V →Codestral 2508 vs Qwen3.5-27B →Codestral 2508 vs MiniMax M2.7 →Codestral 2508 vs Mistral Small 3.1 24B →Any other combination →

Questions, answered

How much does Codestral 2508 cost on Alyph?

Codestral 2508 costs $0.345 per million input tokens and $1.03 per million output tokens on Alyph. A typical task — 5K tokens in, 800 out — works out to about 0.26¢. Billed from your Alyph wallet, with hard spending limits. Prices can change when the provider changes theirs; this page was last checked on August 9, 2026.

What is Codestral 2508's context window?

256K tokens — about 370 pages.

Can Codestral 2508 read images and files?

Codestral 2508 reads Text and Files. Drop a file onto the canvas and it becomes part of the context — the only limit is the context window.

Does Codestral 2508 show its reasoning?

No — Codestral 2508 answers directly, without a visible thinking pass.

Test Codestral 2508 against GLM 4.6V

One prompt, both models at the same time, the better answer wins. Free to start, with $5 of credit.

Sign up free Try the demo