DigestAI news desk

Cut through the AI noise.

Model · multimodal model

YuE2-3B

YuE2-3B is a multimodal model from m-a-p, first covered on 11 September 2026. It is available through an API (licence: CC BY-NC 4.0).

Last updated

Quick facts

Who makes YuE2-3B?
YuE2-3B is made by m-a-p.
When was YuE2-3B released?
The first report of YuE2-3B's release that Digest AI covered was published on 11 September 2026. YuE2-3B Music Model Sets New Benchmark with Editable Scores →
What type of model is YuE2-3B?
YuE2-3B is a multimodal model.
What licence is YuE2-3B released under?
YuE2-3B is released under the CC BY-NC 4.0 licence.
Where is YuE2-3B available?
YuE2-3B is available through an API.

Taken from the coverage, not from our own testing. A question is left out when none of the reports we read gave the answer. All tracked models →

1 story about YuE2-3B

  1. YuE2-3B Music Model Sets New Benchmark with Editable Scores

    YuE2-3B is a newly released open‑source music generation model that outperforms the commercial Suno v5 on the WildSongBench benchmark, achieving a SongBench average of 6.9632 versus Suno’s 6.8721. The model supports…

    1 source primary source
    huggingface.co