One of Japan's largest directories x find the right AI in as little as a minute

▶︎ For those who want to list their service

  1. AI BEST SEARCH
  2. Yue AI
Yue AI

Yue AI
: How to use it, features, and the business problems it solves

Add bookmark

What is Yue AI?

Yue AI refers to "YuE," an open-source AI music-generation model (Apache 2.0, from HKUST/M-A-P and others) that generates full songs with vocals and accompaniment from lyrics, singing in English, Chinese, Japanese, Korean, and more. It's self-hostable from GitHub/Hugging Face and allows commercial use. Note: this page's URL (yueai.ai) is a third-party site offering the model online—separate from the official YuE GitHub project—and its "free" service has no published pricing.

Business problems it solves

About "Yue AI"

What is Yue AI (YuE)

Yue AI refers to "YuE," an open-source AI music-generation model that generates full songs—vocals and accompaniment—from lyrics. YuE is an open-source model (Apache License 2.0) released by HKUST and M-A-P and others, billed as "an open foundation model for full-song generation, similar to Suno.ai." It can sing in multiple languages including English, Chinese, Japanese, and Korean, and you can specify genre and mood.

This page's URL (yueai.ai) is a third-party site that lets you use the open-source YuE model online, and its operator is not disclosed. YuE's official project is on GitHub (multimodal-art-projection/YuE), separate from yueai.ai. This page focuses on the open-source YuE model, which is the substance.

Model
YuE (open source; LLaMA-based two-stage architecture)
Developed by
HKUST / M-A-P and others (official: GitHub multimodal-art-projection/YuE)
Category
Lyrics-to-song (vocals + accompaniment) AI music generation
Languages
English, Chinese (Mandarin/Cantonese), Japanese, Korean, etc.
License
Apache License 2.0 (commercial use allowed; attribution recommended)
How to use
Self-host (GitHub/Hugging Face) or online sites like yueai.ai

Make songs with vocals from lyrics

YuE's trait is that it generates vocals and accompaniment together from lyrics and a genre. With a "dual-track (dual-token)" technique that handles vocals and accompaniment as separate tracks, plus a semantically enhanced audio tokenizer, it aims for natural song structure. Being open source, you can get the model from GitHub or Hugging Face and run it in your own environment.

How to Use (for online sites like yueai.ai)

  1. Open a site

    Go to an online site such as yueai.ai (for self-hosting, follow the GitHub instructions).

  2. Enter lyrics

    Enter the lyrics for the song you want (English, Japanese, etc.).

  3. Choose a style

    Specify a genre or mood, such as pop, rock, or jazz.

  4. Generate and review

    Preview the generated song and regenerate as needed.

Features

01Create

  • Lyrics-to-song — generate a full song with vocals + accompaniment from lyrics
  • Many genres — pop, rock, jazz, hip-hop, and more
  • Multilingual singing — English, Chinese, Japanese, Korean, etc.

02Technology

  • Dual-track generation — generate vocals and accompaniment simultaneously
  • Semantically enhanced tokenizer — aims for high-quality audio output
  • LLaMA-based two-stage architecture — 7B (stage 1) + 1B (stage 2)

03Use

  • Open source / self-host — published on GitHub / Hugging Face
  • Commercial use OK (Apache 2.0) — attribution recommended

The "full songs up to 5 minutes" claim comes from the online site; the official README says "several minutes" and generates short segments by default (longer output needs more generation/compute). Actual length and quality vary by environment and settings.

Pricing

The open-source YuE model is free—you can get it from GitHub/Hugging Face and self-host it (inference needs your own compute). The online site yueai.ai advertises itself as "free," but no pricing page or clear plans/limits are published (its FAQ references a pricing page that does not exist). The previously mentioned "Free (5 songs/mo) / Pro $9.99 / Enterprise" cannot be verified, so it is not listed here. Check [YuE official (GitHub)](https://github.com/multimodal-art-projection/YuE) or the respective sites for the latest (as of August 2026).

How it differs from other AI music generators

YuE's strengths are that it is open source, generates songs with vocals from lyrics, and allows self-hosting and commercial use. For ease and quality, commercial cloud services may suit better.

AspectYuE (Yue AI)SunoMureka
DeliveryOpen source (self-hostable)Cloud serviceCloud service
FocusLyrics → vocals + accompanimentMusic generation broadlyMusic generation, DAW control
PriceThe model is freeFree–subscriptionFree–subscription

To easily make high-quality songs, Suno, Mureka, and AIVA are comparison points. If you want to run an open-source model yourself and use it commercially, YuE is a strong choice.

The information on this page is based on what was publicly available on official sources (GitHub, etc.) as of August 2026. Please check the official sources for the latest specifications and license.

Try Yue AI

Frequently compared services