Live Music Models
Abstract
We introduce a new class of generative models for music called live music models that produce a continuous stream of music in real-time with synchronized user control. We release Magenta RealTime, an open-weights live music model that can be steered using text or audio prompts to control acoustic style. On automatic metrics of music quality, Magenta RealTime outperforms other open-weights music generation models, despite using fewer parameters and offering first-of-its-kind live generation capabilities. We also release Lyria RealTime, an API-based model with extended controls, offering access to our most powerful model with wide prompt coverage. These models demonstrate a new paradigm for AI-assisted music creation that emphasizes human-in-the-loop interaction for live music performance.
Citation
@misc{2025magentart,
title={Live Music Models},
author={Lyria Team and Antoine Caillon and Brian McWilliams and Cassie Tarakajian and Ian Simon and Ilaria Manco and Jesse Engel and Noah Constant and Yunpeng Li and Timo I. Denk and Alberto Lalama and Andrea Agostinelli and Cheng-Zhi Anna Huang and Ethan Manilow and George Brower and Hakan Erdogan and Heidi Lei and Itai Rolnick and Ivan Grishchenko and Manu Orsini and Matej Kastelic and Mauricio Zuluaga and Mauro Verzetti and Michael Dooley and Ondrej Skopek and Rafael Ferrer and Zalán Borsos and Äaron van den Oord and Douglas Eck and Eli Collins and Jason Baldridge and Tom Hume and Chris Donahue and Kehang Han and Adam Roberts},
year={2025},
eprint={2508.04651},
archivePrefix={arXiv},
primaryClass={cs.SD},
url={https://arxiv.org/abs/2508.04651},
}