whatmodel.sh / tts / Zonos v0.1

Zonos v0.1

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Zonos v0.1 is an open-weights text-to-speech model from Zyphra, released February 2025. Its published size is 1.6B parameters. Licence: apache-2.0.

39.1±4.4#89 of 118 text-to-speech modelsfrom 7 benchmark results across 3 sources

Score within the Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot comparison set only.

$ how it compares

The text-to-speech models ranked either side of Zonos v0.1, out of 148 with a score, then more from Zyphra.

#modelscorewhereRAMreleased
112Trendyol-TTS Trendyol39±3comparison only2026-06
113Magpie TTS Multilingual 357M NVIDIA39±6local~0.7GB2026-07
114Zonos v0.1 Zyphra39±4api price unknown +local~4.3GB2025-02
115Maya1 Maya Research38±6local~6.7GB2025-11
116Gemini 3.8 Flash TTS Google37±3api price unknown2026-09
more from Zyphra
63ZONOS2 8B Zyphra57±2local~5.2GB2026-06

see all text-to-speech models · find the ones that fit your machine

$ what it's good at

other tests7 not in a task · 92% of its score
testresultvs the others in its rundifficultycountswhy
EmergentTTS · 6-category preference14.1% ✎▅▅▃▃█▅▅▃·▃·▅ beats 3/16hard · avg 30%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 37%typical
EmergentTTS · v1 paper automatic MOS (WV-MOS)3.39 / 5 ✎▄·█·█▄····█▄ beats 2/8easy · avg 3.93 / 5▇▇▇▇▇▇▇▇ 20%blunt · at its level
EmergentTTS · 6-category WER19.1% ✎▂▄··▂▄▄··█▄▆ beats 5/16avg 16%▇▇▇▇▇▇ 14%blunt
balacoon zero-shot TTS · DAPS-celeb · ReDimNet speaker similarity67.5% ✎▄·······█··▄ beats 2/3easy · avg 71%▇▇▇ 8.4%blunt · at its level
balacoon zero-shot TTS · VCTK · ReDimNet speaker similarity77.5% ✎█···█····█·█ beats 3/3easy · avg 74%▇▇▇ 6.9%blunt · at its level
balacoon zero-shot TTS · VCTK · Whisper-large-v3-turbo CER1.3% ✎█····█····██ beats 1/3avg 1.0%▇ 3.2%blunt
balacoon zero-shot TTS · DAPS-celeb · Whisper-large-v3-turbo CER7.5% ✎██········██ beats 1/3avg 4.1%▇ 3.0%blunt

$ where to run it

Through an API

api · price unknown · zyphra.com ↗

Other ways to run it

  1. Run it as published · needs ~4.34 GB · from github.com/Zyphra (official)
    git clone https://github.com/Zyphra/Zonos && cd Zonos && uv sync && uv pip install -e . && uv run sample.py

Official weights and documentation: huggingface.co/Zyphra ↗

$ how the score was worked out · EmergentTTS-Eval six-category… 51%, balacoon speech_gen_baselines… 22%, EmergentTTS-Eval v1 paper (ar… 20%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EmergentTTS-Eval six-category leaderboard2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 51%Apache-2.0
balacoon speech_gen_baselines · zero-shot TTS metrics (VCTK, DAPS-celeb)4▇▇▇▇▇▇ 22%Apache-2.0
EmergentTTS-Eval v1 paper (arXiv:2505.23009v1) · Table 1 automatic MOS1▇▇▇▇▇▇ 20%CC BY 4.0
prior · pull toward the middle▇▇ 8%
weighted average= 39.1

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json