NeuTTS is the world's first super-realistic on-device text-to-speech speech language model with instant voice cloning, developed by Neuphonic. Built on a compact 0.5B LLM backbone (Qwen 0.5B), it brings natural-sounding speech, real-time performance, and speaker cloning to local devices, unlocking a new category of embedded voice agents, assistants, and compliance-safe applications. The model uses the NeuCodec audio codec that achieves exceptional audio quality at low bitrates using a single codebook.