---
title: "Voxtral Small speech-to-text API hosted in the EU"
description: "Transcribe and understand audio with Voxtral Small 24B from Mistral AI on inference.eu: an OpenAI-compatible API with processing in EU data centres."
canonical: "https://inference.eu/models/voxtral-small-24b/"
language: "en"
---

Speech to text · Mistral AI

# Run Voxtral Small in the EU for transcription and audio understanding

Voxtral Small is Mistral AI's open speech model: it transcribes audio, answers questions about recordings and summarises them, while keeping the text skills of Mistral Small. On inference.eu you call it through an OpenAI-compatible API, and your recordings are processed in EU data centres and never used for training.

[Contact sales](https://inference.eu/contact/)[Example request](#example)

1.  0124B
2.  02EU data centres
3.  03Apache 2.0

## What Voxtral Small is

Voxtral Small 1.0, published as Voxtral-Small-24B-2507 in July 2025, is a 24-billion-parameter model from Mistral AI that adds audio input to the Mistral Small 3 language model. It can run in a dedicated transcription mode, or take audio and text together and answer questions, summarise or follow spoken instructions.

It has a 32K-token context, which Mistral says covers up to about 30 minutes of audio for transcription and 40 minutes for understanding. It detects the spoken language automatically and is strongest in English, Spanish, French, Portuguese, Hindi, German, Dutch and Italian. The weights are released under Apache 2.0.

## What it does well, and where it falls short

Voxtral's advantage over a pure transcription model is that it understands what it hears. You can ask it for a summary of a call, the action items from a meeting or the answer to a question asked in a voice note, in a single request and without a separate language model. It can also trigger function calls directly from spoken requests.

It is a large model for transcription alone, so for high-volume, plain speech to text a smaller dedicated model such as Whisper Large V3 Turbo can be more efficient. Recordings longer than about 30 to 40 minutes need to be split. Mistral notes that system prompts are not yet supported, so put instructions in the user message.

## Calling Voxtral Small through the API

inference.eu uses the OpenAI request format. For plain transcription, send the audio file to the OpenAI-style transcription endpoint with the model set to `voxtral-small-24b`, exactly as you would with OpenAI's speech models after changing the base URL and key.

To ask questions about a recording, send the audio as an `input_audio` content part in a chat message together with your text question. Several audio clips and several turns in one conversation are supported. Mistral recommends `temperature` 0 for transcription and around 0.2 for understanding tasks. Tool definitions work as with any chat model.

## Recordings processed in the EU

Voice recordings often contain names, health details or customer data, and a voice can itself be personal data. With Voxtral Small on inference.eu, audio and transcripts are processed in data centres in the European Union, used only to produce your result and never used to train any model.

Every customer receives a data processing agreement under Article 28 of the GDPR and our list of subprocessors, so you can document how recordings are handled. Billing is in euros. Voxtral Small comes from a French lab and is licensed under Apache 2.0, which allows commercial use without user limits. You remain responsible for informing callers that they are recorded.

## Voxtral Small 24B

Maker

Mistral AI

Made in

France

Type

Speech to text

Parameters

24B

Context window

32,768 tokens

Licence

[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)

Released

July 2025

Model ID

voxtral-small-24b

Processing

EU data centres

Training on your data

Never

Source

[huggingface.co](https://huggingface.co/mistralai/Voxtral-Small-24B-2507)

voxtral-small-24b · Python

```
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.inference.eu/v1", api_key=os.environ["INFERENCE_EU_API_KEY"])

with open("meeting.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(model="voxtral-small-24b", file=audio)
print(transcript.text)
```

## What teams use it for

-   1

    ### Meeting and call summaries

    Turn recorded meetings or support calls into a summary with decisions and action items in one request.

-   2

    ### Voice notes into tasks

    Let field staff record a voice note and have it transcribed, understood and turned into a ticket or a function call.

-   3

    ### Multilingual transcription

    Transcribe interviews and calls in English, French, German, Spanish, Italian, Dutch or Portuguese with automatic language detection.

## Voxtral Small 24B: common questions

The questions buyers, developers and legal teams ask us most. Something missing? Ask us directly.

That depends on how you process recordings. On inference.eu, Voxtral requests are processed in EU data centres, not used for training and covered by a data processing agreement under Article 28 GDPR with a published subprocessor list. You still need a legal basis for recording people and must inform them.

Voxtral Small has a 32K-token context. Mistral states that this covers audio of up to about 30 minutes for transcription and about 40 minutes for understanding tasks. Split longer recordings into parts and combine the transcripts afterwards.

It detects the spoken language automatically. Mistral highlights English, Spanish, French, Portuguese, Hindi, German, Dutch and Italian as the languages where it performs best. Test other languages on your own recordings before relying on them.

Yes. Voxtral Small is released under the Apache 2.0 licence, which allows commercial use, modification and redistribution without user or revenue limits. You can build it into paid products and services.

Whisper Large V3 Turbo is a small model of about 809 million parameters made only for speech recognition and speech translation. Voxtral Small is a 24B model that also understands the audio: it can summarise, answer questions and call functions. Use Whisper for bulk transcription and Voxtral when you need understanding.

Yes. Send the audio together with a text question in a chat message, and Voxtral answers based on what was said. It can also produce structured summaries, so you do not need a separate transcription step followed by a language model.

Yes. Voxtral can trigger functions directly from spoken intent, for example creating a ticket when a caller describes a problem. Define your tools in the standard OpenAI \`tools\` parameter. Mistral describes function calling as supported but still developing.

Change the base URL of your OpenAI client to the inference.eu API, use your inference.eu key and set the model to \`voxtral-small-24b\`. Transcription requests and chat requests with audio keep the same structure.

No. Audio and transcripts sent to inference.eu are used only to produce your result. They are never used to train or fine-tune Voxtral or any other model, and they are not shared with Mistral AI.

According to Mistral's model card, system prompts are not yet supported for Voxtral Small. Put your instructions, such as the summary format or language you want, in the user message alongside the audio instead.

[Ask a question](https://inference.eu/contact/)

## Related

[All models](https://inference.eu/models/#directory)

-   [Whisper Large V3 TurboTranscribe audio with OpenAI's open Whisper Large V3 Turbo on inference.eu: the OpenAI transcription API format, with recordings processed in the EU.Speech to text](https://inference.eu/models/whisper-large-v3-turbo/)
-   [Mistral Small 4 119BRun Mistral Small 4 119B from French lab Mistral AI on inference.eu: reasoning, vision and code in one model, processed in EU data centres.Chat](https://inference.eu/models/mistral-small-4-119b/)
-   [Magistral Small 1.2Run Magistral Small 1.2, Mistral AI's open 24B reasoning model, on inference.eu through an OpenAI-compatible API with processing in EU data centres.Reasoning](https://inference.eu/models/magistral-small/)

## Move your AI workloads to EU infrastructure

Tell us which models you use and roughly how much traffic you have. We reply within one working day with pricing and next steps.

-   Reply within one working day
-   Pricing for your volume
-   DPA ready to sign