Model reference · open weights
MiniMax-Music3 is an open-weight audio or speech model from joemattie, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
MiniMax-Music3 GGUF for audio.cpp Component GGUF package of MiniMaxAI/MiniMax-Music3 for local inference with audio.cpp, the ggml-based audio inference engine. What this is for MiniMax-Music3 generates full songs (vocals plus arrangement, 44.1 kHz stereo, up to six minutes) from a music description caption and lyrics. The official release runs through the diffusers MiniMaxMusic3ModularPipeline in Python; this package exists so the model can run natively on consumer GPUs through audio.cpp's C++/CUDA runtime, with no Python or PyTorch at inference time. The files here were produced while porting MiniMax-Music3 to audio.cpp as the minimaxmusic3 community model family (PR #241). The port reimplements the whole pipeline against these GGUFs: the Qwen3-8B autoregressive stage that emits one semantic code per 40 ms frame with classifier-free guidance, the 4-layer RVQ depth decoder that fills seven residual codebooks per frame, the 36-layer flow-matching transformer that denoises Flow-VAE latents over overlapping 200-frame windows, and the DAC-style vocoder that renders stereo audio. Every component was validated against the diffusers reference implementation (exact tokenizer ids, exact greedy depth-decoder rollouts, 0.9998+ logit correlation on the language model, 0.99998 on the flow transformer); the parity harness lives in tests/minimaxmusic3/ in the audio.cpp tree. This repo is the download source the minimaxmusic3 model spec points at (files are nested under MiniMax-Music3-GGUF/ to match the audio.cpp installer layout), so the native web UI's one-click installer works against it directly. It doubles as the interim distribution point until the package is mirrored into the official audio-cpp/audio.cpp-gguf catalog. Files Conversion choices worth knowing about: - The language model's 200k-row output head is sliced to the 16385 rows the sampler can ever pick (the audio end token plus the 16384 semantic codes), saving about 1.5 GB. - The DiT and depth decoder are stored F16 rather than the checkpoint's BF16: the CUDA BF16 matmul path is several times slower than F16 on Ampere, the weights fit F16's range with a wide margin (max magnitude about 3.2), and component parity
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | joemattie |
|---|---|
| Type | Audio & music |
| Variants | 1 |
| Based on | MiniMaxAI/MiniMax-Music3 |
| Released | 2026-08-14 |
| Popularity | 1k downloads / month |
| Licence | Commercial licence needed |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| MiniMax-Music3-GGUF | — | GGUF | — | — | Weights ↗ |
Using it via the API
Once AxForge deploys joemattie-minimax-music3 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (joemattie-minimax-music3 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="joemattie-minimax-music3" -F file=@audio.mp3
Details
Tags
Licence
The weights are open but its licence needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗