Freedom251 › AI Music Generators
YuE
What YuE is
YuE2 generates songs from lyrics and a style prompt by creating an editable melody-and-chord plan, then rendering vocals and accompaniment. Users can alter harmony, melody, tempo, or form in an ABC score before generating a new recording. For covers, YuE2 can use a transcribed song melody to produce an interpretation in a target style. The hosted NOIZ browser demo accepts song descriptions and lyrics, while its cover studio accepts audio uploads up to 100 MB in formats including MP3, WAV, FLAC, M4A, AAC, and OGG. The listed free hosted demo costs nothing and shows 3 generations left. Local setup specifies Linux, Python 3.12, and an NVIDIA GPU with BF16 support and 24 GB VRAM; output is 48 kHz stereo without quantization. A staged Python API exposes plan(), generate_semantic(), synthesize(), and decode(). Code, documentation, and the agent skill use Apache 2.0; model weights use CC BY-NC 4.0 with additional creator permission.
Who it suits
It suits musicians, content creators, and personal users who want editable song generation, covers, transcription, or score changes. Local users need the specified Linux, Python, and NVIDIA hardware setup; companies need model-weights licensing contact for commercial use.
What it does well
- Editable melody-and-chord planning
- Generates vocals and accompaniment
- Supports cover creation from melodies
- Hosted browser demo available
- Python API and ComfyUI workflow linked
What to know first
- Local use requires Linux
- Local use requires 24 GB VRAM
- Hosted demo shows three generations left
- Commercial company use needs model licensing
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Free hosted demo | NOIZ-hosted browser demo; page shows 3 generations left | Free |
| YuE2 open-source model and hosted demo | local inference requires Linux, Python 3.12 and NVIDIA GPU with BF16 support · online demo requires no installation | Free |
As github.com lists them, read 12 Oct 2026. The maker's own page has the final word.
Verdict
YuE2 combines lyric-driven song creation with editable scores, covers, a browser demo, and local or API workflows. Hardware requirements and model-weight licensing make the deployment choice important for users and companies.
Facts from YuE's own pages
- Academic origin
- The YuE paper describes a family of open foundation models based on LLaMA2 for lyrics-to-song generation of up to five minutes, with style transfer and bidirectional generation. · arxiv.org, 15 Oct 2026
- Agent editing
- The repository provides an agent skill for song generation, instrumental creation, transcription, ABC-score editing and listening comparisons. · github.com, 15 Oct 2026
- Agent workflow
- The repository includes a yue2-music agent skill for song generation, instrumental music, transcription, covers, and score editing. · github.com, 12 Oct 2026
- Audio output
- The documented pipeline produces 48 kHz stereo audio and saves FLAC plus scores, tokens, latents, settings and model identities. · github.com, 15 Oct 2026
- Audio upload
- Yes · github.com, 21 Sept 2026
- Benchmark claim
- The Hugging Face model card reports YuE2 best-of-8 SongBench average 6.9632 on 192 WildSongBench prompts, above the listed Suno v5 and v6 results. · huggingface.co, 15 Oct 2026
- Code licence
- The repository code is released under the Apache License 2.0. · github.com, 15 Oct 2026
- Commercial use
- Yes · github.com, 21 Sept 2026
- Covers
- YuE2 can use a transcribed song melody to create a new interpretation in a target style. · github.com, 12 Oct 2026
- Documentation depth
- The repository includes quick start, creation, cover, editing, agent-skill, benchmark and generation-guide materials with runnable Python and command-line examples. · github.com, 15 Oct 2026
- Editable composition
- The system creates an explicit melody-and-chord plan that users or agents can inspect and change before audio rendering. · github.com, 15 Oct 2026
- Editing
- Users can edit an ABC score’s harmony, melody, tempo, or form and generate a new complete recording. · github.com, 12 Oct 2026
- Free plan
- Yes · github.com, 21 Sept 2026
- Generation API
- The staged Python interface exposes plan(), generate_semantic(), synthesize() and decode() operations. · github.com, 15 Oct 2026
- Hosted demo
- The NOIZ demo accepts song descriptions and lyrics, and its cover studio accepts audio uploads up to 100 MB in listed formats including MP3, WAV, FLAC, M4A, AAC, and OGG. · yue.noizai.net, 12 Oct 2026
- Independent rating availability
- Capterra lists a YuE AI product page with a displayed 0.0 rating and no visible review count in the search result. · capterra.com, 15 Oct 2026
- Inference availability
- Hugging Face states that YuE2-3B is not deployed by any Inference Provider. · huggingface.co, 15 Oct 2026
- Integrations
- Official materials document Hugging Face model loading, ComfyUI native nodes, a NOIZ-hosted demo, Maestro and the SheetSage2 transcription model. · github.com, 15 Oct 2026
- Intended users
- The repository describes personal users, content creators, and musicians as permitted to use YuE2 and monetize generated outputs without fees or royalties payable to the authors. · github.com, 12 Oct 2026
- Licensing
- The repository says code, agent skill, and documentation use Apache 2.0, while model weights use CC BY-NC 4.0 with additional creator permission; commercial use by companies requires contacting the authors about a model-weights license. · github.com, 12 Oct 2026
- Limit: audio quality
- Tom's Guide said low-VRAM versions reduce audio quality and can generate extremely slowly. · tomsguide.com, 15 Oct 2026
- Limit: commercial use
- MrKeyoor notes that the code is Apache-2.0 while the model weights are CC BY-NC 4.0, restricting commercial products that cannot accept noncommercial terms. · mrkeyoor.com, 15 Oct 2026
- Limit: editing behavior
- MrKeyoor reports that editing a score renders a new complete recording and can change singing, timbre or other waveform regions. · mrkeyoor.com, 15 Oct 2026
- Limit: low-VRAM workflows
- Elma says quantized or offloaded configurations can be slower, more fragile and more prone to out-of-memory failures. · elma.sh, 15 Oct 2026
Where YuE stands on Freedom251
- AI Music Generators#1 of 123
- AI Song Generators#1 of 31