Zyphra released ZONOS3-Mini, a compact zero-shot voice-cloning text-to-speech model: a sparse mixture-of-experts with 100M active and 800M total parameters under Apache 2.0, with controls for identity, accent, mood and word-level delivery. It runs in real time on-device (under 150 ms to first audio on an M3 MacBook Air) with runtimes for browsers via WebGPU, SGLang/vLLM and GGML. Weights are on Hugging Face, and it is also served on Zyphra Inference.
Read the official announcement ↗Home / New releases / November 2026 / Week of Nov 9–15, 2026
⚡ ZONOS3-Mini open weights (Zyphra)
Released November 9, 2026 · Verified against the official source · Official announcement (zyphra.com) ↗
Also released the week of Nov 9–15, 2026
- 2026-11-13 Personal Agent Protocol (Poppy) v1.0 (Sierra)
- 2026-11-13 Full-suite app API: web search, files, OCR, ASR and TTS (DeepSeek)
- 2026-11-13 llama.cpp v1.0.0 and ggml v1.0.0 (ggml-org)
- 2026-11-13 HeyGen Decision 1 (decision model API)
- 2026-11-12 Gemini 4 Hydrogen (Google)
- 2026-11-12 Claude Meetings (Anthropic)
- 2026-11-12 Inkling 2 Small open weights (Thinking Machines Lab)
- 2026-11-12 Atlas world model opens to everyone (World Labs)
Frequently asked questions
When was ZONOS3-Mini open weights (Zyphra) released?
ZONOS3-Mini open weights (Zyphra) was released on November 9, 2026, according to the official announcement (zyphra.com). SIloadout added it to its verified release archive for the week of Nov 9–15, 2026.
Where is the official announcement for ZONOS3-Mini open weights (Zyphra)?
The official source is https://www.zyphra.com/our-work/zonos3-mini. Zyphra released ZONOS3-Mini, a compact zero-shot voice-cloning text-to-speech model: a sparse mixture-of-experts with 100M active and 800M total parameters under Apache 2.0, with controls for identity, accent, mood and word-level delivery.
← Newer: Emma and Nuplex Beta API (Nuance Labs) · Older: Seedream 5.5 (ByteDance) → · Week of Nov 9–15, 2026 · November 2026 · Open-weight releases · All releases · JSON Feed