July 15, 2025:
Mistral Unveils Voxtral: Open AI Audio Breakthrough - French AI startup Mistral has introduced Voxtral, an open-source audio model for businesses, aiming to challenge proprietary systems with affordable, high-quality speech intelligence. Built on Mistral Small 3.1 LLM, Voxtral can transcribe and understand audio in multiple languages for up to 40 minutes, aiding in real-time actions and summaries.
Voxtral is available in two variants, Small and Mini, and includes a cost-effective transcription-focused API. Users can test Voxtral for free via an API on Hugging Face or through Mistral's chatbot, Le Chat, as the company continues to advocate for open-source AI solutions.
