YuYuE2 logo
Audio & Voice

YuE2

AI tool that generates full songs from lyrics and style prompts.

At a glance

Price not confirmed · Source review 2026-10-24
Price & billing

Open weights free for non-commercial use; self-hosting costs are your own GPU/compute. Third-party hosted studios sell credit packs separately. Checked October 2026.

Free allowance

Open weights free for non-commercial use; self-hosting costs are your own GPU/compute.

Best suited to

Researchers, developers, and technically skilled musicians who want inspectable, editable AI composition and can self-host on a 24GB GPU.

Fit is editorial guidance, not a measured result.
Platforms

Access: not separately verified. Operating systems not separately verified.

Published-source facts and editorial guidance. No product test is implied. Confirm current conditions before purchase.

No product screenshot is claimed for this listing. Use the official website ↗ to inspect the current interface.

What to know before choosing

Good fit

Researchers, developers, and technically skilled musicians who want inspectable, editable AI composition and can self-host on a 24GB GPU.

Limits and trade-offs

Weights are non-commercial (CC BY-NC 4.0); self-hosting needs significant GPU resources; published benchmark scores come from the lab's own evaluation protocol.

Plans and free allowances

Open weights free for non-commercial use; self-hosting costs are your own GPU/compute. Third-party hosted studios sell credit packs separately. Checked October 2026.

Evidence

Official website ↗ (reviewed 2026-10-24)

Official published documentation reviewed; not a product test

Dates: Last reviewed 2026-10-24. Last reviewed is the latest source or editorial review date; it is not a content-change date.

A closer look at YuE2

YuE2 is an open-weights AI music model from Multimodal Art Projection that turns lyrics and a style prompt into full songs with vocals, uniquely writing an editable symbolic score (melody and chords) before rendering 48 kHz stereo audio.

Editorial guidance based on published documentation. The workflow below is a suggested starting point, not a measured product result.

When to choose YuE2

  • Researchers, developers, and technically skilled musicians who want inspectable, editable AI composition and can self-host on a 24GB GPU.
  • Musicians who want an editable symbolic score (melody and chords) written before high-quality stereo rendering.

When to choose something else

  • Non-technical users wanting a polished app
  • commercial projects — the weights are licensed CC BY-NC 4.0 (non-commercial).

A practical starting workflow

  1. Download the YuE2-3B weights and set up the inference environment (Linux, Python, NVIDIA GPU with ~24GB VRAM).
  2. Provide lyrics and a style prompt, then review and optionally edit the generated ABC-format score.
  3. Render the song to 48 kHz stereo audio, and iterate using the chat-based or agentic editing modes.

Check before committing

  • Weights are non-commercial (CC BY-NC 4.0); self-hosting needs significant GPU resources; published benchmark scores come from the lab's own evaluation protocol.
  • Pricing and feature availability may change; confirm current terms on the official website before committing.

Common questions

Is YuE2 really open source?

The code is Apache-2.0 and the weights are published, but the weights carry a CC BY-NC 4.0 non-commercial license — free for research and personal use, not for commercial products.

Official website ↗

What hardware do I need to run YuE2?

The official quick start specifies a 24GB NVIDIA GPU with BF16 support, Linux, and Python 3.10+.

Official website ↗

What makes YuE2 different from Suno or Udio?

YuE2 writes an editable symbolic score (melody and chords in ABC notation) before rendering audio, so you can change the composition itself rather than re-rolling generations.

Official website ↗

Compare approaches for this task

  • Suno — 需要把想法快速变成完整歌曲的人:短视频创作者找背景音乐或主题曲、独立音乐人做 demo 小样、播客/品牌方定制片头曲、给朋友写生日歌的普通用户。尤其是需要导出音频文件、发到 Spotify/YouTube 等平台并商用的人,Suno 是目前同类里链路最完整的(下载、分轨、商用授权都可用)。
  • Udio — 把 AI 作曲当灵感实验和打磨练习的人:想逐段雕琢一首歌的词曲爱好者、喜欢对比两个版本再选的创作者、想体验 Voices(可复用的人声风格)和风格库混搭的玩家。免费版每天能做 3 首完整长度的歌,不绑卡,适合先免费玩明白再决定。
  • Stable Audio — Producers, sound designers and developers who need instrumental beds, loops or SFX

Sources and checks

Last source review is not a date of product testing. No rating or test result is implied by inclusion. Check the current plan and terms before buying.