>
K
Model Details
Introduction
Kokoro is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient. With Apache-licensed weights, Kokoro can be deployed anywhere from production environments to personal projects.
Benchmarking
FLOPs
168000000000000000000Notes: 312000000000000 FLOP / GPU / sec * 500 GPU - hours * 3600 sec / hour * 0.3 [assumed utilization] = 1.6848e+20 FLOP
Training
Training Code Accessibility
Apache 2.0 for weights
https://huggingface.co/hexgrad/Kokoro-82M
Apache 2.0 for inference code
https://github.com/hexgrad/kokoroHardware
NVIDIA A100 SXM4 80 GBSize Notes: "Few hundred hrs" of training audio data
Parameters
Parameters
82000000Notes: 82M
