Downloads · 30 days
0
nerualdreming/open-rvq-encoder-minimax-music3-169m-v4-onnx
open-rvq-encoder-minimax-music3-169m-v4-onnx is a audio-to-audio model from nerualdreming. Use it for the audio-to-audio task on the model card, and read the license before you ship it in a product. It is set up for onnx. The card lists the license as apache-2.0.
MiniMax Music 3 writes music from a description and lyrics. What it cannot do out of the box is read music: the model was released without the encoder that turns audio into the codes it generates, so a finished track…
Downloads · 30 days
0
Access
Public
Updated Aug 19, 2026
Repo size
676 MB
Likes
0
Public
Click a slice to open those files.
.onnx676 MB · 100%
From the Hugging Face model README
MiniMax Music 3 writes music from a description and lyrics. What it cannot do out of the box is read music: the model was released without the encoder that turns audio into the codes it generates, so a finished track cannot be handed back to it.
That one missing piece is the difference between a slot machine and an editor. Without it, a chorus you dislike means generating the whole song again and losing everything you liked. With it:
It is not a style or voice reference. Identity is decided upstream in the language model; handing it audio says "work with this material", not "imitate this artist".
The encoder was reconstructed by the community and published as PyTorch weights. This is the 169M v4 checkpoint exported to ONNX, so it runs anywhere ONNX Runtime runs — on the card or the processor, with no Python and no PyTorch.
Verified against the PyTorch reference on random input: every one of the eight codes per frame identical, not merely close.
| Source weights | SimpleTuner/open-rvq-encoder-minimax-music3-169m-v4, final/ |
| Parameters | 169M, f32, 676 MB |
| Exported by | scripts/export-rvq-encoder-onnx.py |
| Runs in | MiniMax Music3 Studio |
| name | shape | meaning |
|---|---|---|
latents | [1, T, 128] | DAC-VAE latents at 86.13 Hz — what neural-codec writes from a WAV or MP3 |
pool | [1, F, T] | pooling matrix onto the model's 25 Hz frame grid |
codes | [1, F, 8] | per frame: one semantic code of 16384, then seven acoustic codes of 1024 |
Those eight codes per frame are exactly what the engine accepts as
audio_codes, which skips the autoregressive stage and renders from them.
track.mp3
-> neural-codec (audio -> VAE latents, 86.13 Hz) [in minimaxmusic.cpp]
-> this model (latents -> 8 codes per frame) [ONNX Runtime]
-> mm-server audio_codes (codes -> audio) [in minimaxmusic.cpp]
Read from the checkpoint, not assumed: conv_in (kernel 7, 128 → 1088), three
residual blocks with dilations 1/3/9, pooling onto the 25 Hz grid, eight
transformer layers (17 heads, d_model 1088, muP attention scale 8.0), a
semantic head over 16384 entries, then a two-layer depth decoder (d 512) that
produces the seven acoustic codes in a greedy chain.
The reconstruction is the community's work — SimpleTuner for the weights and the reference implementation, and the discussion around MiniMaxAI/MiniMax-Music3 for working out that it was possible at all. This repository only converts it to a format that runs without PyTorch.