Voxtral 4B TTS 2603

JSON →
mistral audio
audio

A 4-billion-parameter text-to-speech model by Mistral AI for multilingual speech synthesis.

context window 8K tokens
max output 8K tokens
streamingfunction-calling