m-a-p

/

YuE2-3B

Like 1.21k

Follow

Multimodal Art Projection 1.61k

Dear Multimodal Art Projection (M-A-P) team,

I am writing to request clarification regarding the licensing terms of YuE2-3B, specifically concerning the commercial use of audio generated with the model.

The Hugging Face model card lists the weights under CC BY-NC 4.0, and the MODEL_LICENSE file in the GitHub repository grants rights for NonCommercial purposes only. However, neither document states explicitly whether the restriction extends to the output generated by the model (i.e., the resulting songs, vocals and instrumentals), or applies solely to the model weights and their redistribution or adaptation.

I would therefore be grateful if you could confirm the following points:

1. May audio generated with YuE2-3B (including YuE2-Vae) be used for commercial purposes, for example monetization on platforms such as YouTube, Facebook or Spotify?
2. May such audio be sold or licensed to third parties, or used in paid client work?
3. If commercial use is not permitted under the current license, is a separate commercial license available, and what are the terms and procedure for obtaining it?
4. If commercial use of the output is permitted, are there any attribution requirements or other conditions I should observe?

A written confirmation would be very helpful, as I would like to ensure that my use of the model is fully compliant with your license terms.

Thank you very much for your time and for making YuE2 available to the community.

Kind regards,
Vito