3B
Parameters
Dense Transformer
AIVory Smart Inference · Voxtral
Open WeightVoxtral Mini 3B - voice input, pocket size.
Mistral's 3B voice model. Accepts audio input natively. The smallest model in the Smart Inference catalogue, priced accordingly.
At a glance
Voxtral Mini 3B via Smart Inference.
Routed to the cheapest live provider on every request. Same model, lower bill.
32K
Context window
tokens per request
—
Cheapest input price
per 1M tokens via Smart Inference
—%
Cheaper
vs the priciest endpoint for the same model
Pricing
Voxtral Mini 3B: price comparison.
Our cheapest routed price vs what you'd pay elsewhere.
Prices in USD per 1M input tokens · updated live from the router
Quick start
Use Voxtral Mini 3B in three lines.
OpenAI-compatible. Swap your base_url, keep everything else.
from openai import OpenAI
client = OpenAI(
base_url="https://smart.aivory.net/v1",
api_key="sk-aivory-...",
)
resp = client.chat.completions.create(
model="voxtral-mini-3b-2507",
messages=[{"role": "user", "content": "Ship it."}],
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://smart.aivory.net/v1",
apiKey: process.env.AIVORY_API_KEY,
});
const resp = await client.chat.completions.create({
model: "voxtral-mini-3b-2507",
messages: [{ role: "user", content: "Ship it." }],
});
console.log(resp.choices[0].message.content);curl https://smart.aivory.net/v1/chat/completions \
-H "Authorization: Bearer $AIVORY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "voxtral-mini-3b-2507",
"messages": [{"role": "user", "content": "Ship it."}]
}'That's the whole migration. Get early access and try it →
Use cases
What Voxtral Mini 3B is great at.
-
Voice-Activated Commands
Accept spoken input from users and respond with text. No separate speech-to-text pipeline. Voxtral Mini understands the speech and processes the request in one API call. Good for simple voice commands, quick questions, and voice-to-text conversion.
-
Mobile and Edge Voice Apps
At 3B parameters, Voxtral Mini is small enough to self-host on minimal infrastructure. Pair it with a mobile app that records audio and sends it to a lightweight backend.
-
Voice Intake for Larger Models
Use Voxtral Mini to transcribe and understand voice input, then pass the text to a larger model (Llama 3.3, Qwen 2.5) for complex processing. Keeps voice handling cheap while using a bigger model only for the hard part.
FAQ
Common questions.
What does Voxtral Mini 3B cost on AIVory?
The cheapest model in the catalogue by parameter count. Check the live table above for current rates.
How do I send audio to the API?
Pass base64-encoded audio in the messages array alongside any text prompt. The API follows the OpenAI-compatible multimodal format. Check the API reference for the exact schema.
Voxtral Mini vs Voxtral Small - which for voice?
Mini (3B) is faster and cheaper. Small (24B) understands complex speech better, handles more languages, and produces more accurate transcription. Use Mini for simple voice commands, Small for conversations and multilingual audio.
Can Voxtral Mini replace Whisper?
For simple transcription, it's comparable. Whisper is still better for noisy audio, heavy accents, and pure transcription accuracy. Voxtral Mini's advantage is that it understands and responds, not just transcribes.
Start using Voxtral Mini 3B. Today.
Swap your base_url, keep shipping.
The router is live. Credits from $10, no subscription, no commitment.