Submit a model

Know an open-weight TTS or STT model we're missing, or have measured numbers for a device we don't cover? Send the Hugging Face or GitHub link (and your method) and we'll verify it, usually within a week.

Open a submission (placeholder link)

Model authors: want a featured listing or an earlier benchmark? Featured items are labelled, and rankings and benchmark numbers can never be bought.

New open speech models: will they run on your hardware?

One email a week: new open TTS/STT models, license gotchas, and what we measured on CPUs, Macs, Pis and consumer GPUs.