---
title: YuE2-3B-bf16
canonical_url: "https://www.modelscope.cn/models/mlx-community/YuE2-3B-bf16"
md_url: "https://www.modelscope.cn/models/mlx-community/YuE2-3B-bf16.md"
repository: mlx-community/YuE2-3B-bf16
last_updated: 2026-10-28
license: cc-by-nc-4.0
model_type:
  - yue2
architectures:
  - YuE2ForCausalLM
base_model:
  - m-a-p/YuE2-3B
base_model_relation: finetune
parameters: 3.6B
tensor_type:
  - BF16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 18
stars: 0
tags:
  - mlx
  - swift
  - music-generation
  - yue2
  - non-commercial
---

# YuE2-3B-bf16

> YuE2-3B-bf16 - mlx-community 在 ModelScope 开源的模型。Non-commercial weights (CC BY-NC 4.0). Songs made with this model may not be sold, licensed or used in paid work, except under M-A-P's individual-creator permission (MODELLICENSE, 2026-09-16): personal users,…

mlx-community/YuE2-3B-bf16 是 ModelScope 魔搭社区上的 3.6B 参数机器学习模型，采用 cc-by-nc-4.0 许可，基于 m-a-p/YuE2-3B 构建。

- **Repository**: mlx-community/YuE2-3B-bf16
- **License**: cc-by-nc-4.0
- **Parameters**: 3.6B
- **Base model**: m-a-p/YuE2-3B
- **Tags**: mlx, swift, music-generation, yue2, non-commercial
- **Downloads**: 18
- **Stars**: 0
- **Last updated**: 2026-10-28

Source: https://www.modelscope.cn/models/mlx-community/YuE2-3B-bf16

---

# YuE2-3B-bf16 (MLX)

> **Non-commercial weights (CC BY-NC 4.0).** Songs made with this model may not be sold, licensed or used in paid
> work, **except** under M-A-P's *individual-creator permission* (MODEL_LICENSE, 2026-09-16): personal users, content
> creators and musicians acting in an individual capacity may publish, sell and license songs they make with it.
> Companies may not use the model commercially without a licence from M-A-P (ryuanab@connect.ust.hk), and a hosted
> service making songs for paying users is commercial use. Responsible use is a condition of the permission. Covers
> and supplied lyrics may need the original song's rights; this licence doesn't grant them.

The backbone of [m-a-p/YuE2-3B](https://huggingface.co/m-a-p/YuE2-3B) (M-A-P; code
[multimodal-art-projection/YuE](https://github.com/multimodal-art-projection/YuE)) for
[mlx-yue2-swift](https://github.com/xocialize/mlx-yue2-swift), a Swift-MLX port for Apple Silicon. YuE2 turns
lyrics and a style prompt into a song with vocals: it plans a chord-annotated ABC score, then sings it (48 kHz
stereo, up to 6 minutes). The score is editable and can be supplied back to re-render.

This repository is the 3.58 B Mixture-of-Transformers backbone. Its tensors are bf16 and bit-equal to the release.
The VAE decoder is [mlx-community/YuE2-Vae-fp32](https://huggingface.co/mlx-community/YuE2-Vae-fp32).

| file | size |
|---|---|
| `model.safetensors` (628 tensors, bf16) | 7.26 GB |
| `qwen.tiktoken` (tokenizer, as shipped by m-a-p) | 2.6 MB |
| `config.json` (+ an `mlx_yue2` section: dtype, source pin, generation defaults) | 1.6 KB |
| `LICENSE` (HF, CC BY-NC 4.0), `MODEL_LICENSE` (GitHub, with the individual-creator permission), `NOTICE`, `THIRD_PARTY_NOTICES.md`, `licenses/` | — |

## Use (Swift, MLXEngine)

```swift
// Package.swift: .package(url: "https://github.com/xocialize/mlx-yue2-swift", from: "0.1.0") → product "MLXYuE2"
import MLXServeCore, MLXToolKit, MLXYuE2

let engine = MLXServeEngine()   // CC-BY-NC-4.0 is outside the permissive policy: it registers with a licence advisory
let id = try await engine.register(YuE2Package.registration,
                                   configuration: YuE2Configuration(tier: .bf16, modelsRootDirectory: modelsRoot),
                                   id: "yue2-3b")
try await engine.prepare(.music, package: id)   // downloads this repo and YuE2-Vae-fp32 on first use
let song = try await engine.run(MusicRequest(caption: "English, warm piano pop, expressive female voice, 88 BPM",
                                             lyrics: "[Verse]\n…\n\n[Chorus]\n…", seed: 831001), package: id) as! MusicResponse
// song.audio: 48 kHz stereo WAV; song.score: the ABC plan (edit it, pass it back as MusicRequest.score)
```

Tiers are quantised at load from this bf16 repo: `.bf16`, `.ar8` (the AR path 8-bit), `.int8` (the AR and NAR paths
8-bit). Memory (phys_footprint at the 6-minute worst case, M5 Max): bf16 7.6 GB resident / 14.6 GB peak, ar8 6.4 / 13.3,
int8 5.1 / 11.9. An English 58 s song renders in 41 s at bf16 on an M5 Max. Planning modes `full` and `off`; `melody`
is not offered.

## Parity

The Swift port is gated against an fp32 PyTorch oracle of the release: tokenizer id-exact; AR logits argmax 100 %
(relL2 ≤ 1.8e-6); NAR velocity ≤ 1.6e-5; VAE ≤ 4.6e-6; end-to-end audio ≤ 1.7e-4. bf16 and 8-bit tiers sit 1.2–1.6e-2
from fp32 in velocity, where the release itself runs. Details:
[MEASUREMENTS.md](https://github.com/xocialize/mlx-yue2-swift/blob/main/MEASUREMENTS.md).

## Attribution

YuE2, YuE2-3B, https://huggingface.co/m-a-p/YuE2-3B, by M-A-P, under CC BY-NC 4.0. The weights here are unmodified
apart from serialisation; `NOTICE` lists exactly what changed. Not endorsed by M-A-P.