Voxtral Mini 4B Realtime

Hot

Mistral's multilingual real-time speech transcription model with configurable latency (240ms-2.4s), supports 13 languages

Model Specifications

ArchitectureAUDIO

Parameters4B

Familyvoxtral

VRAM (Q4)4GB

4B ASR model, <500ms delay matching offline transcription accuracy. Apache 2.0 license

#mistral#stt#speech-to-text#realtime#trendingSource

Send this model's specs directly to your community.

Fast speech-to-text, 13 languages

High-accuracy multilingual transcription

Mistral's open-weight text-to-speech model with 9-language support, voice cloning from 3s of audio, and 70ms latency

A complete breakdown of quantization levels and VRAM overhead for running local models.

Comparing NVIDIA and AMD options for the best speed-to-dollar ratio.

Understanding local AI formats and which one to pick for your specific hardware.