Kokoro-82M: tiny open-source TTS model runs on CPU with quality of far larger models
solyarisoftware · x · 2026-09-28
- Kokoro-82M by hexgrad is an open-weight TTS model with just 82M parameters, Apache-2.0 licensed, that runs comfortably on CPU while delivering audio quality comparable to much larger models.
- v1.0 (Jan 2025) was trained on only a few hundred hours of data using 1,000 A100 GPU hours; API market pricing is under $1 per million input characters (under $0.06 per hour of audio output).
- The author warns that third-party sites with "kokoro" in their domain (e.g. kokorottsaicom) are likely scams unaffiliated with the project.
More from Infra
- Training-free page-sparse decode in llama.cpp: 1.72× faster 128k decoding on an RTX 5090 — topppggg · 2026-11-16
- AMD MI455 beats Nvidia Rubin on raw specs, claims viral chip debate — zephyr_z9 · 2026-11-16
- llama.cpp offloads CI caches and runners to Hugging Face to survive 35 PRs/day — ggerganov · 2026-11-16
- India's Sarvam launches Token Factory, hosting top open models on local infra — HarveenChadha · 2026-11-16
- ZGX Fury stress-tested: 351 tok/s on Qwen3.8-Flash-Next, offline agents and local RAG demos — Sam Witteveen · 2026-11-16
- Gemini 3.5 Flash now runs on air-gapped Google Distributed Cloud, offline up to 12 months — shashib · 2026-11-16