How to Use Voxtral Small 24B in Instinct
Use Voxtral Small 24B for speech transcription, translation, and questions about recordings. Its 32k context supports up to 30 minutes of transcription or 40 minutes of audio understanding.
Prompts to try with Voxtral Small 24B in Instinct
Copy an example into Instinct, or ask in your own words.
Every morning, use Voxtral Small 24B to summarize yesterday's support recordings, each under 30 minutes. Report the issue and next step per call.
Each Friday, use Voxtral Small 24B to summarize this week's meeting recordings individually. Compile stated decisions and unresolved questions into a weekly digest.
Each evening, use Voxtral Small 24B to transcribe new interview recordings under 30 minutes. Preserve the original language and flag unclear passages.
Every Monday, use Voxtral Small 24B to translate new Spanish customer interview excerpts into English. Keep each interview's findings separate.
After each training session, use Voxtral Small 24B to summarize its recording in sections under 30 minutes. Produce key lessons and review questions.
How to set up Voxtral Small 24B in Instinct
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Instinct this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use Spiral's MCP server
- Create an API key
In the console, open API keys and create a key for Instinct. Copy it.
- Send Instinct this message
Instinct reads Spiral's skill and asks for your API key.
set up https://spiral.ai/SKILL.md
- Give Instinct the key
When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Voxtral Small 24B
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "mistralai/voxtral-small-24b-2507", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Instinct finds Voxtral Small 24B on Spiral, shows the price and runs it.
Voxtral Small 24B providers and prices
Spiral runs Voxtral Small 24B through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.006148 per run at the default inputs. $0.13 per 1M input tokens, $130.00 per 1M audio input tokens, $0.39 per 1M output tokens.
Settings you can ask for
Ask Instinct for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 26214 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 |
Every API field of Mistral: Voxtral Small 24B 2507 is on its tool page: open its card above.
What can you do with Voxtral Small 24B in Instinct?
Build recurring written digests of recorded speech with Voxtral Small 24B. Give each run a specific extraction task and collect the results into a reviewable report.
Meeting decision notes
Summarize a meeting recording into decisions, assigned actions, and unresolved questions. Request names and deadlines only when explicitly stated.
Interview transcripts
Turn a recorded interview of up to 30 minutes into a transcript for editorial review.
Translated interview excerpts
Translate a Spanish, French, German, or Italian interview excerpt into English text for an editor.
Support call summaries
Extract the customer's problem, troubleshooting attempted, and agreed next step from a recorded support call.
Lecture study notes
Summarize a lecture segment, then ask questions about the explanation and examples it contains.
Podcast research
Pull an episode segment's main arguments and supporting examples into a research brief.
Spoken product feedback
Turn a recorded user interview into reported problems, requested changes, and follow-up questions.
More on Voxtral Small 24B and Instinct
Overview
Voxtral Small 24B in other agents
More Instinct guides
Questions about Voxtral Small 24B in Instinct
Can Instinct use Voxtral Small 24B?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Voxtral Small 24B like any built-in tool.
How much does Voxtral Small 24B cost in Instinct?
Up to $0.006148 per run at the default inputs. $0.13 per 1M input tokens, $130.00 per 1M audio input tokens, $0.39 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.
Do I need an OpenRouter account to use Voxtral Small 24B in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Voxtral Small 24B on its own OpenRouter account.
How long does it take to set up Voxtral Small 24B in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Voxtral Small 24B runs?
Voxtral Small 24B runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.
What is Voxtral Small 24B useful for?
Transcribing speech, translating spoken content, summarizing recordings, and answering questions about what was said. Use a focused question when you need a particular detail from a recording.
How long can a recording be?
Mistral documents up to 30 minutes for transcription and 40 minutes for audio understanding, with a 32k token context window. Split longer recordings into shorter sections before requesting a combined summary.
Which languages does Voxtral Small 24B support?
Mistral lists English, Spanish, French, Portuguese, Hindi, German, Dutch, and Italian. Specify whether you want the original language preserved or the content translated.
Can Voxtral Small 24B generate a spoken reply?
It produces text from audio or text input. Choose it for transcripts and written answers; speech generation requires a separate text-to-speech model.
What sampling settings should I start with?
Mistral recommends temperature 0.2 and top-p 0.95 for audio understanding. For transcription, its recommendation is temperature 0.
Start using Voxtral Small 24B in Instinct
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Instinct:
set up https://spiral.ai/SKILL.md


