Zyphra/ZONOS3-Mini
Can I trust it?
No red flags in our checks.
Stars from real accounts No data
We haven't checked its stars for fakes yet.
Active OK
Last change 2 days ago.
Licence OK
MIT: you can use it for free, also in a commercial product. Keep the licence notice.
Quick checks from public data. They help, but they are no guarantee.
What it is
Inference runtime for the ZONOS3-Mini multilingual text-to-speech model. Provides HTTP servers for CUDA, MLX, ggml (CPU, CUDA, Metal, Vulkan) and an in-browser WebGPU build, with zero-shot voice cloning from a reference clip and an offline Python API.
Who it's for
Developers running local or self-hosted text-to-speech with voice cloning on NVIDIA GPUs, Apple Silicon or CPU
How to use it
Install
uv sync --locked --all-packagesFrom its README. Check it there before running it.
Ask your AI to set it up
Paste this into Claude, ChatGPT or your coding assistant.
Help me install and set up Zyphra/ZONOS3-Mini in my project. Follow its README: https://github.com/Zyphra/ZONOS3-Mini. Explain each step and check it is safe before running anything.
Star curve
Get alerts for repos like this
We'll email you when something in Models & inference starts rising. Free, once a week, only what matches.
What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy
For maintainers
Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.
[](https://repoalerts.com/repo/Zyphra/ZONOS3-Mini?ref=badge)Similar repos
- Edge0-AI/Audio8_TTS
Compact text-to-speech models with zero-shot voice cloning, in 0.6B and 0.1B sizes. The 0.6B Realistic release supports 23 languages; the 0.1B model targets Chinese and English. Includes Python infere
- plotarmordev/drama-tensorfold
Recipe that builds the TensorFold native Zig/CUDA engine with opt-in patches to run Fish Audio's Drama 3 text-to-speech on an NVIDIA GB10. It starts a local server with an OpenAI-style /v1/audio/speec
- zxcHolmes/minimax-h3-studio
A self-hosted web studio for MiniMax-H3 audio-video generation on a single NVIDIA GPU machine. It drives ComfyUI as a Python library and provides a resident job server and browser UI for text, image a
- n4ze3m/tiny-voca
A 4.5M-parameter English text-to-speech model with two voices (F2 female, M2 male) at 24 kHz. Trained from scratch, it runs on CPU, in PyTorch, ONNX Runtime, or in the browser, and can be exported to
- inclusionAI/Ming-3.0-tts
Multilingual text-to-speech models (0.5b dense and tiny MoE) with voice cloning, voice design, built-in voices, natural-language style instructions and inline paralinguistic tags. Includes inference c
- debpalash/VoiceStudio
Desktop application for local voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation, with a local API and MCP server for agents. Runs on NVIDIA CUDA, Apple Silico