# Voxtral: what I've logged

> Voxtral comes up in 3 posts of my AI log, from February 2026 to October 2026, and each item below links to the primary source. The sources come from Mistral AI and Mistral, filed under frontier, open weights and image, video, world.

## Items

1. **Voxtral Mini 4B Realtime Arabic** (Mistral AI, 5 Oct 2026): An open streaming speech-to-text model for Modern Standard Arabic and Gulf, Levantine, Nile Valley and Maghrebi dialects, built to handle code-switching. [Primary source](https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-Arabic) · [Post](https://heyaliasif.com/ai/2026-10-05-week-in-ai#item-8)
2. **Voxtral TTS** (Mistral AI, 23 Mar 2026): Mistral's first text-to-speech model, with open weights and low latency. [Primary source](https://mistral.ai/news/voxtral-tts) · [Post](https://heyaliasif.com/ai/2026-03-23-week-in-ai#item-2)
3. **Voxtral Transcribe 2** (Mistral, 2 Feb 2026): A new speech-to-text family from Mistral, including a real-time model with open weights. [Primary source](https://mistral.ai/news/voxtral-transcribe-2) · [Post](https://heyaliasif.com/ai/2026-02-02-week-in-ai#item-5)

## Posts

- 2026-10-05 [Mistral Large 4, Claude Haiku 5.5 and a wave of decision models](https://heyaliasif.com/ai/2026-10-05-week-in-ai)
- 2026-03-23 [Google's TurboQuant, open voice from Mistral and Cohere, Flash Live](https://heyaliasif.com/ai/2026-03-23-week-in-ai)
- 2026-02-02 [Claude Opus 4.6 and GPT-5.3-Codex land on the same day](https://heyaliasif.com/ai/2026-02-02-week-in-ai)

Canonical: https://heyaliasif.com/ai/topics/voxtral