MiniMax-Music3 flow matching DiT needs about 2.1 GB of memory at Q4_K_M (a 1.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 13 quants, from 0.1 GB (F32) to 17.2 GB (BF16).
MiniMax-Music3 flow matching DiT – 2.4B parameter music generation
MiniMax-Music3 flow matching DiT is a 2.4B parameter model that generates full songs up to five minutes. It uses a Flow Matching DiT to render latent tracks, producing 32 kHz, 16‑bit stereo WAV audio. The model is part of a hierarchical system with a global LLM and a local acoustic LLM.
Good at
- Full‑song generation up to five minutes
- Structured caption and lyric control
- High‑fidelity 32 kHz audio output
Limits
- No vision support
- Requires structured caption input
- Large VRAM for higher quants
Best for
- Music production pipelines
- Songwriting assistants
- Audio research projects
Who it’s for. Developers and musicians looking to generate complete songs with detailed control over style, vocals, and arrangement. The model is suitable for integration into web or desktop applications that support GGUF inference.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F32 | 0.1 GB | 0.6 GB | Fits | Get file → | |
| F32 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| Q8_0 | 0.7 GB | 1.3 GB | Fits | Get file → | |
| BF16 | 1.3 GB | 1.9 GB | Fits | Get file → | |
| Q4_K_M | 1.4 GB | 2.1 GB | Fits | Get file → | |
| Q5_K_M | 1.7 GB | 2.4 GB | Fits | Get file → | |
| Q6_K | 2.0 GB | 2.8 GB | Fits | Get file → | |
| Q8_0 | 2.6 GB | 3.4 GB | Fits | Get file → | |
| Q5_K_M | 6.3 GB | 7.5 GB | Tight | Get file → | |
| Q6_K | 7.0 GB | 8.4 GB | Too big | Get file → | |
| Q8_0 | 9.1 GB | 10.7 GB | Too big | Get file → | |
| F32 | 9.7 GB | 11.4 GB | Too big | Get file → | |
| BF16 | 17.2 GB | 19.7 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (file size 1.39 GB, memory 2.1 GB at 8k context). For 12 GB VRAM, Q5_K_M (1.69 GB, 2.4 GB). For 16 GB VRAM, Q6_K (2.01 GB, 2.8 GB). Larger VRAM allows Q8_0 (0.69 GB, 1.3 GB) for higher quality.
What GPU do you need for MiniMax-Music3 flow matching DiT?
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 Q8_0 (2.6 GB file, about 3.4 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (2.6 GB file, about 3.4 GB in use).
- 8 GB VRAM: the best quant that fits is Q5_K_M (6.3 GB file, about 7.5 GB in use).
- 12 GB VRAM: the best quant that fits is F32 (9.7 GB file, about 11.4 GB in use).
- 16 GB VRAM: the best quant that fits is F32 (9.7 GB file, about 11.4 GB in use).
- 24 GB VRAM: the best quant that fits is BF16 (17.2 GB file, about 19.7 GB in use).
Which MiniMax-Music3 flow matching DiT quant should you choose?
Q4_K_M is the usual starting point for MiniMax-Music3 flow matching DiT: it balances file size and quality well. Here it is a 1.4 GB file.
The Q8_0 file is 0.7 GB, about 0.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is F32 at 0.1 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 MiniMax-Music3 flow matching DiT
How much VRAM does MiniMax-Music3 flow matching DiT need?
About 2.1 GB at Q4_K_M with an 8K context window. That is the 1.4 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 MiniMax-Music3 flow matching DiT on an 8 GB GPU?
Yes, with Q5_K_M: the 6.3 GB file needs about 7.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of MiniMax-Music3 flow matching DiT GGUF?
It depends on the quant: F32 is 0.1 GB, F32 is 0.3 GB, Q8_0 is 0.7 GB, BF16 is 1.3 GB, Q4_K_M is 1.4 GB, Q5_K_M is 1.7 GB and 7 more.
Where can I download MiniMax-Music3 flow matching DiT GGUF?
The files are on Hugging Face in the Serveurperso/MiniMax-Music3-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 MiniMax-Music3 flow matching DiT use?
Hugging Face lists the license as other. 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 MiniMax-Music3 on 8 GB VRAM?
Yes, using the Q4_K_M quantization which requires about 2.1 GB of VRAM at 8k context.
Does the model support GPU inference?
Yes, it runs on CUDA, Vulkan, and CPU via minimaxmusic.cpp.
What audio format does it output?
It outputs 32 kHz, 16‑bit stereo WAV audio.