# SGLang-Omni adds day-0 support for Zonos3-Mini voice-cloning TTS

> Original title: Zonos3-Mini from @ZyphraAI runs on SGLang-Omni on day 0!

- Source: SGLang
- Published: 2026-11-09T22:22:15.000Z
- Bad Influence: https://badinfluence.news/items/t43nqdnazshfcs0olen9ytrmh
- Original: https://x.com/sgl_project/status/2119918516212318260

## Summary

SGLang-Omni now runs Zyphra's Zonos3-Mini on day 0, according to SGLang. The 800M-parameter MoE text-to-speech model activates 80M parameters per token. On one H100, first audio streams back in 57 ms.

## Post

Zonos3-Mini from @ZyphraAI runs on SGLang-Omni on day 0!

It’s an 800M MoE voice-cloning TTS model with 80M params active per token. On one H100, first audio streams back in 57 ms.

Serve it today with SGLang-Omni: https://sgl-project.github.io/sglang-omni/cookbook/zonos3.html

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWt2ctNbAAA37nB.png&mode=thumb&exp=1794528000&sig=7b3b88a63c32c922)

## Quoted post @ZyphraAI

> Meet ZONOS3-Mini: open-source text-to-speech with voice cloning and word-level control. All on-device and super fast.
> 
> Every voice in this video is from a TTS model that can run real-time on a phone.
> 
> Comment “ZONOS” within 2 days for 20 hours of free audio.
> 
> Sound on 🔊

https://x.com/ZyphraAI/status/2119914895861076151