README.md

metadata

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
  - 8bit

MiniMax Music 3 · MLX 8-bit

Native MLX affine 8-bit weights for MiniMaxAI/MiniMax-Music3, converted for lyric-conditioned song generation on Apple Silicon with mlx-audio.

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 · 6-bit · 4-bit · MXFP8 · MXFP4 (experimental) · 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-Music3-8bit \
  --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-Music3-8bit")
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

• Affine 8-bit weight quantization with group size 64.
• 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.
• The three weight shards total 14,167,688,378 bytes.
• Strict loading and real generation produced finite 44.1 kHz stereo audio.
• 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.