---
title: "YuE2 Now Generates Full Songs Locally Without Python or PyTorch"
subtitle: "Pre-quantized GGUF weights and a portable C++ backend let YuE2 generate 48kHz stereo songs on a single consumer GPU with an editable ABC score."
url: "https://alphasignal.ai/news/yue2-now-generates-full-songs-locally-without-python-or-pytorch"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-09-30T00:57:12.006Z"
updated: "2026-09-30T01:00:23.251Z"
type: "Model"
topics:
  - "Audio"
  - "Open Source"
subtopics:
  - "Music Generation"
upvotes: 3465
source_url: "https://huggingface.co/Serveurperso/YuE2-GGUF"
comments: 0
truncated: true
---

# YuE2 Now Generates Full Songs Locally Without Python or PyTorch

> Pre-quantized GGUF weights and a portable C++ backend let YuE2 generate 48kHz stereo songs on a single consumer GPU with an editable ABC score.

# YuE2 brings local song generation to GGUF and C++

YuE2 can now generate complete songs through a native C++17 stack. The release combines [YuE2 GGUF weights](https://huggingface.co/Serveurperso/YuE2-GGUF) with the [yue2.cpp runtime](https://github.com/ServeurpersoCom/yue2.cpp), built on GGML for CPU, CUDA, and Vulkan. Users provide style tags and lyrics, and the pipeline returns 48 kHz stereo audio alongside the ABC score it composed.

## YuE2 plans before it renders

Multimodal Art Projection released YuE2-3B as an open-weight model for generating songs with vocals and accompaniment. Its authors report results competitive with Suno v5 and v6 on WildSongBench, although that claim is benchmark-specific and comes from the model team.

YuE2 first writes melody and chord information in ABC, a plain-text notation format, then generates the audio representation. That intermediate score gives developers an editable checkpoint between the prompt and the final track. The GGUF port preserves this process while replacing the Python and PyTorch runtime with native executables.

## Three weight sets drive the pipeline

The release provides three model families converted from the upstream checkpoints:

| Component    | Role                                                                                                     | Available sizes                                                                                         |
| ------------ | -------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- |
| **Backbone** | A 3.6B-parameter Mixture-of-Transformers that generates the score, semantic codes, and acoustic latents. | 7.17 GB in BF16 to 2.62 GB in Q5\_K\_M. The download script selects the 3.81 GB Q8\_0 build by default. |
| **VAE**      | An Oobleck SnakeBeta decoder that converts acoustic latents into stereo audio.                           |                                                                                                         |

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [Serveurperso/YuE2-GGUF on Hugging Face](https://huggingface.co/Serveurperso/YuE2-GGUF)
- [yue2.cpp on GitHub](https://github.com/ServeurpersoCom/yue2.cpp)
- [YuE2-3B on Hugging Face (MAP)](https://huggingface.co/m-a-p/YuE2-3B)
- [YuE2-3B in ComfyUI overview](https://comfyui-wiki.com/en/news/2026-09-11-yue2-3b)
- [YuE2 paper (arXiv 2503.08638)](https://arxiv.org/html/2609.33757)