Making Kokoro TTS 50-100x faster without knowing how
retained text
My thinking was models, esp Opus 5.5 can just hill-climb as long as you give them something achievable and a clear metric to improve on. So here's something I don't even have a clue how it exactly works. The training itself happened in my basement 4070ti super over a day or so. And yes I know its Opus slop, but still really cool how things I never thought possible unless you're an ML engineer is available to a run of the mill web-dev guy. Thanks to this tweet from Sahil for even showing how the initial idea worked to distill Kokoro in the first place. [https://x.com/sahilmdkr/status/2112212041045889409](https://x.com/sahilmdkr/status/2112212041045889409)
document fetched 2026-10-23 00:46 · canonical https://reddit.com/r/LocalLLaMA/comments/1xcuqor/making_kokoro_tts_50100x_faster_without_knowing