Skip to model information

Voxtral Small

Available

A Mistral AI model with a 32K-token context window and tool calling. Available on the Mistral API.

Access details: Mistral API

An alias. The provider can point it at a newer version.

voxtral-small-latest
Updated Sep 23, 2026

Capabilities & limits

Context window
32K
Maximum output
Not published
Image input
Unsupported
Tool calling
Supported
Structured output
Supported

Pricing

USD / 1M tokens
Input
$0.10
Output
$0.40

Standard rates per million tokens. Discounts and long-context rates, where offered, are on Mistral AI's pricing page.

Full rate card 3 rates

All prices in USD.

Audio input

$0.004 per minute

Text input

$0.10 per 1M tokens

Text output

$0.40 per 1M tokens

From the model page; the docs pricing page does not list Voxtral Small. Batch processing is billed at a 50% discount.

Price history
Sep 23, 2026

Input $0.10 per million tokens

Sep 23, 2026

Output $0.40 per million tokens

Technical details

Parameters
24B
Active parameters
24B
Training / knowledge cutoff
Not published
Input modalities
Audio, text
Output modalities
Text

Release & lifecycle

First available
—
available
As of Sep 23, 2026

Considering a switch?

Specs only go so far. Try both on your own tasks and compare quality, speed and cost.

Compare with GPT-6 Luna →

Prices and availability as of October 20, 2026. Mistral AI's own pages have the latest.