YuE2: 12GB+ VRAM, ComfyUI Paths
- 1. At a Glance
- 2. Overview
- 3. Specifications
- 4. Performance and Quality
- 5. Strengths and Use Cases
- 6. Differences from Similar Models
- 7. Hardware Requirements
- 8. Can You Run It Locally?
- 9. Our Own Measurements
- 10. Distributed Files
- 11. How to Get It
- 12. Related Articles
- 13. What to Read Next
- 14. Sources
- 15. Update History
At a Glance
| Item | Value |
|---|---|
| Repository | Comfy-Org/YuE2 |
| Family guide | YuE2 guide (3 articles) |
| Publisher guide | M-A-P: models and licenses |
| Published | 2026-09-11 |
| License | cc-by-nc-4.0 |
| Formats | safetensors |
| Source type | Primary source (the publisher itself) |
Values determined by this site’s code when the information was collected. Dates are JST.
Overview
Comfy-Org has released Comfy-Org/YuE2, a set of model files repackaging the open music generation model “YuE2-3B" for execution within ComfyUI. It is a text-to-audio model that generates songs with vocals and accompaniment from lyrics and style prompts, allowing control over melodies and chords via editable scores.
Specifications
- Parameter Count: 3.6B (based on specifications of the base model
m-a-p/YuE2-3B) - Architecture: AR-NAR Mixture-of-Transformers (
YuE2ForCausalLM) - Output Specification: 48 kHz stereo audio
- Supported Languages: Chinese (zh), English (en)
- License: cc-by-nc-4.0 (non-commercial use only)
Performance and Quality
Evaluation results on WildSongBench and SHS100K are published in the base model’s model card. Below are the evaluation results for full song generation on WildSongBench.
→ Scroll horizontally to see all columns
| Model | Musicality ↑ | SongBench Avg ↑ | MuLan ↑ | AllMusicCaps ↑ | Q3O ↑ | PER ↓ |
|---|---|---|---|---|---|---|
| YuE 1 | 4.0847 | 4.9165 | 0.2623 | 0.2882 | 3.7301 | 36.38% |
| SongBloom | 3.4493 | 4.2350 | 0.2697 | 0.1926 | 3.0287 | 19.19% |
| LeVo 2 | 5.4590 | 6.3247 | 0.3542 | 0.2680 | 3.9458 | 26.12% |
| ACE-Step 1.5 | 5.1588 | 6.0118 | 0.4372 | 0.3869 | 4.5809 | 7.46% |
| HeartMuLa | 5.4963 | 6.2483 | 0.3823 | 0.2786 | 3.4907 | 10.71% |
| DiffRhythm 2 | 4.4775 | 5.2428 | 0.3782 | 0.3255 | 4.0870 | 18.41% |
| Muse | 5.1692 | 6.0349 | 0.3937 | 0.3466 | 4.4038 | 33.42% |
| MiniMax Music 3 | 5.3482 | 6.2830 | 0.3928 | 0.3609 | 4.4362 | 6.27% |
| YuE2 | 5.9075 | 6.7316 | 0.5068 | 0.4054 | 4.6819 | 8.44% |
| YuE2 (best-of-8) | 6.2666 | 6.9632 | 0.5051 | 0.3980 | 4.7009 | 9.79% |
The base model YuE2 records the highest SongBench average among the evaluated open and proprietary models. On the other hand, other models excel in certain metrics, demonstrating a balance between generation quality and prompt compliance.
Strengths and Use Cases
In addition to generating music from lyrics and style prompts, the base model supports melody and chord planning, score editing via ABC notation, and agent-based song editing and revision. It supports Chinese and English lyrics, and is intended for zero-shot cover song generation and text-based music production.
Differences from Similar Models
While Music Generation Model “YuE2-3B" Released, Producing Songs from Lyrics and Style covered the standalone release of the base model “YuE2-3B", Comfy-Org/YuE2 covered in this article repackages those model files to fit the directory structure for ComfyUI.
Hardware Requirements
Estimated requirements (calculated by Local Model Watch) — 3.6B parameters (taken from the base model m-a-p/YuE2-3B)
| Your VRAM | Quantization | File size | Est. memory needed |
|---|---|---|---|
| 12GB (RTX 4070 / 3060 12GB, etc.) | BF16 | 6.8GB | 8.1GB |
Memory estimates add a 20% runtime overhead (KV cache, etc.) to the actual size of the distributed files. Actual usage varies with context length, batch size and inference engine. These figures are computed by this site from file sizes, not published by the model’s authors. Compare with other models in our VRAM quick reference. What the quantization names mean: glossary.
Can You Run It Locally?
The publisher distributes this model as safetensors.
License — cc-by-nc-4.0 (Commercial use prohibited):
Commercial use is prohibited (NC = NonCommercial). Internal business use can also count as commercial, so avoid this license for work use.
Compiled by this site’s code from the published formats and the license field. License summaries are not legal advice — check the publisher’s original terms before relying on them.
Our Own Measurements
Measurements We Did Not Take
This model’s license (cc-by-nc-4.0) does not permit commercial use. Because this site carries advertising, we did not run the model (no CPU run, answers, quantization comparison, conversion or generation).
Distributed Files
Weight files published in Comfy-Org/YuE2, listed by this site from the Hugging Face API. Sizes are the actual file sizes.
| File | Size | ComfyUI folder |
|---|---|---|
audio_encoders/sheetsage2_bf16.safetensors |
1.39GB | models/audio_encoders/ |
checkpoints/yue2_3b_bf16.safetensors |
7.80GB | models/checkpoints/ |
checkpoints/yue2_3b_int8_convrot.safetensors |
3.96GB | models/checkpoints/ |
The ComfyUI folder is shown only where the repository’s layout mirrors ComfyUI’s models/ directory.
How to Get It
Available from Hugging Face at Comfy-Org/YuE2. Place the files according to the following folder structure:
📂 ComfyUI/
└── 📂 models/
├── 📂 audio_encoders/
│ └── sheetsage2_bf16.safetensors
└── 📂 checkpoints/
└── yue2_3b_bf16.safetensors
Related Articles
- yue2-mothersuperior-realaudio-tokenizer-v4: 12GB+ VRAM, File List
- YuE2-3B Music Generation Model for Lyrics and Style: 12GB+ VRAM
- Ming-Image UI Design-Specialized Image Generation Model: ComfyUI Paths
- ComfyUI v0.36.0 Released: Generic Loops, Yue2 & Marigold v2
What to Read Next
- Find models by VRAM (This model runs from the 12GB tier) → Other models that run on a 12GB GPU
- Explore the same model family → YuE2 family overview (3 articles)
- Engines that run this model → ComfyUI
- Formats this model is available in → Safetensors format guide and models
- Learn about the publisher → M-A-P: models, licenses and articles
Sources
Update History
- 2026-09-24: Rebuilt the article: Added a “can you run it locally?" verdict from the distribution formats, converted builds, engine registries, license terms and measured compression
- 2026-09-25: Changed the title to show what the article covers (VRAM requirements, file list, etc.).