Model registry / Mistral AI / Voxtral / Voxtral Mini
Voxtral Mini
Speech transcription model for accurate audio-to-text and captioning workflows
Line
Voxtral
Weights
API only
Released
2026-02-01
Context
not stated
Input
not stated
Output
not stated
Max output
not stated
Coverage
not covered yet
Overview
Voxtral Mini is a speech transcription model from Mistral AI, built for accurate audio-to-text and captioning workflows. It is part of the voxtral line, following the earlier Voxtral Small release. The context window and maximum output are not stated. The model is not open weights, does not support reasoning or tool calling, and accepts audio input to return text output. Pricing is not stated.
Specs
Pricing
Same headline figures as Voxtral Small.
Strengths
- Accurate speech transcription
- Built for captioning workflows
- Accepts audio input
- Returns text output
Best for
- Reach for it for transcribing audio to text
- Reach for it for generating captions
- Reach for it for audio-to-text workflows
How to access
1 host serve this model at the price above · the maker's documentation
Voxtral: every version
The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.
FAQ
- What is the context window?
- The context window is not stated for Voxtral Mini.
- Does it support tool calling or reasoning?
- No. Voxtral Mini does not support tool calling or reasoning mode.
- Is Voxtral Mini open weights?
- No, the weights are not open.