Can Voxtral Small 24B run locally?

Yes, the weights are open and available on Hugging Face. Mistral recommends running it with vLLM, and you will need approximately 55 GB of GPU RAM in bf16 or fp16.

What does Voxtral Small 24B cost to use via API?

Text is priced at $0.10 per million input tokens and $0.30 per million output tokens. Audio input is billed separately at $100 per million seconds.

How is Voxtral Small different from Mistral Small 3.1?

Voxtral Small is built directly on Mistral Small 3.1 and keeps all its text capabilities, but adds a dedicated audio input layer for transcription, translation, audio Q&A, and voice-triggered function calling.

Voxtral Small 24B 2507

5768°

#llm #mistralai #audio #multilingual #local #budget

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Mistral | 32,768 token context | $0.10/M input, $0.30/M output. Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio is priced at $100 per million seconds.

Guides for this model

How much VRAM do I actually need for local AI?

Open weights vs open source vs API-only: what you are actually allowed to do

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Voxtral Small 24B 2507 is Mistral's 24-billion-parameter audio-capable model, built on the Mistral Small 3.1 backbone with a dedicated audio understanding layer added on top. It handles speech transcription, translation, and audio Q&A in a single model, automatically detecting the source language across eight major languages. Open weights and a low API price make it a practical choice for developers who want to avoid stitching together separate ASR and language model pipelines.

Known for

Dedicated transcription mode that auto-detects source language without configuration.
Handles audio up to 30 minutes for transcription, 40 minutes for understanding.
Function-calling triggered directly from spoken voice input.
Natively multilingual: English, Spanish, French, Portuguese, Hindi, German, Dutch, Italian.
Retains the full text capabilities of Mistral Small 3.1 alongside audio.

Watch out for

System prompts are not yet supported, which limits some deployment patterns.
Running locally requires roughly 55 GB of GPU RAM in bf16 or fp16.
Audio pricing is separate from text: $100 per million seconds of input audio.

Best for

Prompts to try

Transcribe and summarize audio

Attach a voice recording of a meeting or lecture. Transcribe everything said, then produce a structured summary with the main topics covered, any decisions made, and action items mentioned. Present the summary as a numbered list.

Audio Q&A from a recording

Attach an audio clip of a spoken explanation or interview. Answer the following questions based only on what is said in the recording: What is the main argument being made? What evidence or examples are given? Are there any caveats or qualifications the speaker mentions?

Translate spoken audio to English

Attach an audio file where the speaker is using a language other than English. Identify the language being spoken, transcribe the original, and provide a clean English translation. Note any phrases that were ambiguous or difficult to render directly.

Compare with

Mistral Nemo
Open-weights model
131,072 token context · $0.03/M input, $0.03/M output
Ministral 3 3B 2512
Open-weights model
131,072 token context · $0.10/M input, $0.10/M output
Ministral 3 14B 2512
Open-weights model
262,144 token context · $0.20/M input, $0.20/M output
Mistral Small 3
Open-weights model
32,768 token context · $0.05/M input, $0.08/M output

More text models from mistralai

Questions people ask

Can Voxtral Small 24B run locally?: Yes, the weights are open and available on Hugging Face. Mistral recommends running it with vLLM, and you will need approximately 55 GB of GPU RAM in bf16 or fp16.
What does Voxtral Small 24B cost to use via API?: Text is priced at $0.10 per million input tokens and $0.30 per million output tokens. Audio input is billed separately at $100 per million seconds.
How is Voxtral Small different from Mistral Small 3.1?: Voxtral Small is built directly on Mistral Small 3.1 and keeps all its text capabilities, but adds a dedicated audio input layer for transcription, translation, audio Q&A, and voice-triggered function calling.

Captain's verdict
A rare single model that does serious transcription and audio understanding without sacrificing text quality, at a price that does not sting.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Voxtral Small 24B 2507 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm mistralai audio multilingual local budget agents

Browse by Model Family