Kitten TTS 2 on BestTTS

This BestTTS page is an independent reference for Kitten TTS 2, a 1.7-billion-parameter speech language model from KittenML (checked 2026-10-22 against publisher documentation). BestTTS is a free directory of speech models; it does not provide the model. It covers ternary transformer weights (about 1.58 bits), reference voice cloning from a 5–30 second recording, 123 voices and 16 languages with named voices per the model card, CPU execution including a C++ runtime, getting-started steps (pip install kittenml) and the Stellon Labs Community License, which requires registration for commercial use and is not unrestricted open source. No real-time benchmark is claimed.