library_name: mlx-audio
pipeline_tag: text-to-audio
base_model: MiniMaxAI/MiniMax-Music3
license: other
license_name: minimax-music3-community-license
license_link: https://huggingface.co/MiniMaxAI/MiniMax-Music3/blob/main/LICENSE
inference: false
tags:
- mlx
- mlx-audio
- apple-silicon
- music-generation
- text-to-music
- text-to-audio
- quantized
- mxfp4
MiniMax Music 3 路 MLX MXFP4 (experimental)
Native MLX MXFP4 weights for
MiniMaxAI/MiniMax-Music3,
converted for lyric-conditioned song generation on Apple Silicon with
mlx-audio. This is an experimental
memory-saving variant; use MXFP8 when lyric fidelity matters.
Community conversion, not an official MiniMax release. All model credit goes to MiniMax. Review the original model card and license before use.
Other MLX variants:
BF16
路 8-bit
路 6-bit
路 4-bit
路 MXFP8 (recommended)
路 NVFP4 (experimental)
Install
MiniMax Music 3 support was merged upstream in
Blaizzy/mlx-audio#888.
Until a PyPI release includes it, install the upstream merge commit directly:
python -m pip install "mlx-audio @ git+https://github.com/Blaizzy/mlx-audio.git@784b29e2691a93ca7483147d86f61859dfaa6296"
Generate
python -m mlx_audio.music.generate \
--model mlx-community/MiniMax-Music3-mxfp4 \
--caption "Warm acoustic pop, 96 BPM, intimate female vocal" \
--lyrics $'[verse]\nMorning light across the room\n[chorus]\nSing with me' \
--duration 30 \
--steps 30 \
--seed 7 \
--output song.wav
from mlx_audio.music import load
model = load("mlx-community/MiniMax-Music3-mxfp4")
result = next(
model.generate(
text="Warm acoustic pop, 96 BPM, intimate female vocal",
lyrics="[verse]\nMorning light across the room\n[chorus]\nSing with me",
duration=30,
steps=30,
seed=7,
)
)
print(result.audio.shape, result.sample_rate) # stereo, 44100 Hz
Lyrics are required by the checkpoint contract. Use [instrumental] explicitly
for instrumental generation. Duration is a requested upper bound: the
autoregressive stage may emit its end token early. Style, tempo, instrument, and
vocal controls are probabilistic rather than strict.
Quantization and verification
- MXFP4, group size 32: E2M1 linear weights with E8M0 group scales.
- Large linears in the global language model, RVQ depth decoder, and flow transformer are quantized. Embeddings, output heads, convolutions, condition encoder, and vocoder remain dense for fidelity.
- Approximately 8.3 GB on disk, versus 27 GB for the BF16 conversion.
- Strict loading and real generation produced finite 44.1 kHz stereo audio.
- This variant showed weaker lyric adherence than BF16 and MXFP8 in controlled checks across two seeds: words may be altered, omitted, or replaced. Treat it as a memory/quality tradeoff rather than an equal-fidelity release.
- The mlx-audio regression suite passed 1,742 tests with 34 expected skips; the focused music, converter, and registry suite passed 43 tests and 3 subtests.
Converted with mlx-audio 0.4.8 development commit c2fa486 and MLX 0.31.2.
License
The weights remain subject to the
MiniMax-Music3 Community License,
including its acceptable-use and commercial terms. The full license text is
included in this repository.