YuE

YuE is an open foundation model from HKUST and the Multimodal Art Projection (M-A-P) research community that turns lyrics into full-length songs with vocals and accompaniment, across genres and several languages. It was one of the first open models to handle long-form songs with singing, and it's released under the Apache 2.0 license, which allows commercial use. It needs a strong GPU to run locally, though ComfyUI wrappers and hosted versions make it easier. It's a common alternative to ACE-Step for people building on open models.