All models
lightweight

Kokoro-82M

Apache-2.0

82M params, Apache 2.0, runs on a potato

Kokoro is an 82-million-parameter text-to-speech model that consistently punches far above its weight, topping community TTS leaderboards against models many times its size. It ships with a set of polished preset voices and is fast enough to run in real time on modest hardware — which makes it even faster behind a hosted API.

Voices

8 voices
Female
Belladefault
american
Nicole
american
Emma
british
Sky
american
Male
Michael
american
Adam
american
George
british
Onyx
american

The scripts

Neutral
“The quick brown fox jumps over the lazy dog near the riverbank.”
Emotional
“I can't believe you actually did it. This is incredible!”
Numbers & Dates
“On March 14th, 2025, the team raised $4.2 million at a 38% margin.”

Install

pip install kokoro

Kokoro-82M FAQ

Do I need a GPU to run Kokoro-82M?

For practical speeds, yes — plan on roughly 1 GB of VRAM. If you'd rather not manage a GPU, hosted access to every model in this directory is coming via OpenSpeech Cloud.

Can Kokoro-82M clone voices?

No — Kokoro-82M uses preset voices. Browse the voice-cloning category for models that clone from reference audio.

What license is Kokoro-82M released under?

Kokoro-82M is released under the Apache-2.0 license. Always check the repository for the exact terms — some models license code and weights separately.

What languages does Kokoro-82M support?

Kokoro-82M supports English, Japanese, Chinese, Spanish, French, Hindi, Italian, Portuguese.

Is there a hosted Kokoro-82M API?

Yes — OpenSpeech Cloud is launching hosted, pay-per-minute access to Kokoro-82M and every other model in this directory. No CUDA, Docker, or GPU needed.