SongGeneration v2 large needs about 8.3 GB of memory at Q6_K (a 6.9 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 2 quants, from 6.9 GB (Q6_K) to 7.6 GB (Q8_0).
SongGeneration v2 large – a 5.12B‑parameter Qwen model for music generation
SongGeneration v2 large is a 5.12B‑parameter model based on the Qwen architecture. It is released under the MIT license and is available in GGUF format. The model is designed for generating music and related audio content.
Good at
- Music composition
- Melody generation
- Audio lyric writing
Limits
- No vision support
- Limited context length
- Requires 8.3–9GB VRAM for 8k context
Best for
- Creating original song drafts
- Generating instrumental loops
- Assisting songwriters
Who it’s for. Developers and musicians looking for a lightweight, open‑source music generation tool. The model is suitable for local deployment on machines with 8GB or more VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q6_K | 6.9 GB | 8.3 GB | Too big | Get file → | |
| Q8_0 | 7.6 GB | 9.0 GB | Too big | Get file → |
For 8GB VRAM use the Q6_K quant (6.94GB file, ~8.3GB memory). If you have 9GB or more, Q8_0 (7.62GB file, ~9GB memory) offers higher precision. Memory estimates are approximate.
What GPU do you need for SongGeneration v2 large?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q8_0 (7.6 GB file, about 9.0 GB in use).
Which SongGeneration v2 large quant should you choose?
Q6_K (6.9 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SongGeneration v2 large
How much VRAM does SongGeneration v2 large need?
About 8.3 GB at Q6_K with an 8K context window. That is the 6.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run SongGeneration v2 large on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of SongGeneration v2 large GGUF?
It depends on the quant: Q6_K is 6.9 GB, Q8_0 is 7.6 GB.
Where can I download SongGeneration v2 large GGUF?
The files are on Hugging Face in the smthem/SongGeneration-v2-large-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does SongGeneration v2 large use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run SongGeneration v2 large on 8GB VRAM?
Yes, the Q6_K quant uses about 8.3GB VRAM at 8k context, which fits on an 8GB GPU with some headroom.
What license does this model use?
The model is released under the MIT license, allowing free use and modification.
Does this model support image or video input?
No, the model has no vision capabilities and only processes text prompts.