Comet Lab Atlas

A small, open-weight text-to-speech model that produces natural voices on modest hardware.

Kokoro is a compact open-weight text-to-speech model. Despite its small size it produces natural, pleasant speech, and it runs comfortably without a heavy GPU.

Its Apache license and modest footprint make it the practical open choice when speech synthesis has to stay local.

Where it's ideally used

A fit when you need good-quality, self-hosted text-to-speech on limited hardware, with a permissive license.

Where it doesn't fit

Not the match for the most expressive, emotionally nuanced output — a hosted premium voice still leads there.