MOSS Music 8B Instruct needs about 10.3 GB of memory at Q8_0 (a 8.7 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 3 quants, from 1.7 GB (F16) to 16.4 GB (F16).
MOSS Music 8B Instruct – local music understanding model
MOSS Music 8B Instruct is an open‑source model for music analysis. It can transcribe lyrics, caption music, identify chords, tempo, key, and structure. The model is available in GGUF format for local inference.
Good at
- Lyrics ASR with timestamps
- Musical captioning and tagging
- Chord, key, tempo reasoning
Limits
- Requires audio tower file, not usable with llama.cpp alone
- No vision support
- Large memory for Q8_0 quant
- No chain‑of‑thought reasoning
Best for
- Local music captioning
- Audio‑to‑text transcription
- Music feature extraction
Who it’s for. Developers building local music analysis tools or datasets. Researchers needing a fully local, open‑source model for music understanding without API calls.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F16 | 1.7 GB | 2.4 GB | Fits | Get file → | |
| Q8_0 | 8.7 GB | 10.3 GB | Too big | Get file → | |
| F16 | 16.4 GB | 18.9 GB | Too big | Get file → |
For 8 GB VRAM, use Q8_0 quant (file size 8.71 GB, estimated 10.3 GB memory at 8 k context). For 4 GB VRAM, use F16 quant (file size 1.72 GB, estimated 2.4 GB memory). The F16 quant is the reference model and offers the best quality.
What GPU do you need for MOSS Music 8B Instruct?
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: the best quant that fits is F16 (1.7 GB file, about 2.4 GB in use).
- 6 GB VRAM: the best quant that fits is F16 (1.7 GB file, about 2.4 GB in use).
- 8 GB VRAM: the best quant that fits is F16 (1.7 GB file, about 2.4 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (8.7 GB file, about 10.3 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (8.7 GB file, about 10.3 GB in use).
- 24 GB VRAM: the best quant that fits is F16 (16.4 GB file, about 18.9 GB in use).
Which MOSS Music 8B Instruct quant should you choose?
Q8_0 (8.7 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is F16 at 1.7 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about MOSS Music 8B Instruct
How much VRAM does MOSS Music 8B Instruct need?
About 10.3 GB at Q8_0 with an 8K context window. That is the 8.7 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 MOSS Music 8B Instruct on an 8 GB GPU?
Yes, with F16: the 1.7 GB file needs about 2.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MOSS Music 8B Instruct GGUF?
It depends on the quant: F16 is 1.7 GB, Q8_0 is 8.7 GB, F16 is 16.4 GB.
Where can I download MOSS Music 8B Instruct GGUF?
The files are on Hugging Face in the scragnog/MOSS-Music-8B-Instruct-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 MOSS Music 8B Instruct use?
Hugging Face lists the license as apache-2.0. 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 MOSS Music 8B Instruct on 8 GB VRAM?
Yes, the Q8_0 quant requires about 10.3 GB memory at 8 k context, which is close to 8 GB VRAM. It may need swapping or lower context.
Does this model support vision tasks?
No, it is a music understanding model and does not support vision.
Is the model available for commercial use?
The model is licensed under Apache‑2.0, which permits commercial use.