Voxtral Mini

Parameters

4.7B

Published

Jul 15, 2025

Voxtral Mini is an AI model developed by Mistral AI (France), first published in July 2025. It works in the speech and language domain, on tasks such as audio question answering, speech recognition (asr), speech-to-text and 2 more.

Epoch AI has no training-compute estimate for this model. The model has 4,700,000,000 parameters.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of Ministral 3B. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Mistral AI
Country of organization: France
Domain: Speech, Language
Task: Audio question answering, Speech recognition (ASR), Speech-to-text, Language modeling/generation, Question answering
Parameters: 4,700,000,000
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: Ministral 3B
Epoch confidence: Confident