File size: 2,125 Bytes
71a08a4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 | ---
library_name: mlx
license: apache-2.0
license_link: https://huggingface.co/OpenMOSS-Team/MOSS-SoundEffect-v2.0/blob/main/README.md
pipeline_tag: text-to-audio
base_model: OpenMOSS-Team/MOSS-SoundEffect-v2.0
tags:
- mlx
---
# mlx-community/MOSS-SoundEffect-v2.0-bf16
This model [mlx-community/MOSS-SoundEffect-v2.0-bf16](https://huggingface.co/mlx-community/MOSS-SoundEffect-v2.0-bf16)
was converted to MLX format from
[OpenMOSS-Team/MOSS-SoundEffect-v2.0](https://huggingface.co/OpenMOSS-Team/MOSS-SoundEffect-v2.0)
— a text-to-sound-effect diffusion pipeline (foley / ambience / creature /
action audio, 48 kHz, up to 30 s) with a 1.3B Wan-style flow-matching DiT, a
continuous 128-d DAC VAE (50 Hz latents), and a frozen Qwen3-1.7B text encoder.
Precision: DiT bf16, DAC-VAE fp32 (the reference decodes under fp32 autocast), Qwen3 text encoder bf16.
## Use with mlx
```bash
pip install moss-sfx-mlx # https://github.com/xocialize/moss-soundeffect-mlx
```
```python
from moss_sfx_mlx.pipeline_mlx import MossSoundEffectPipeline
pipe = MossSoundEffectPipeline.from_pretrained("mlx-community/MOSS-SoundEffect-v2.0-bf16")
audio = pipe(prompt="a heavy wooden door creaks open slowly",
seconds=5, num_inference_steps=100, cfg_scale=4.0, seed=0)
# audio: (1, 1, samples) mx.array at 48 kHz
```
## Parity
Validated against the upstream PyTorch reference (fp32, CPU stream, per-module
and end-to-end golden tensors; full suite in the GitHub repo):
- End-to-end waveform vs PyTorch golden (10-step CFG denoise): max_abs < 1e-2 fp32
- Full-DiT velocity field at production scale (T=1500): max_abs < 1e-2 fp32
- DAC-VAE decode vs reference: max_abs < 1e-2 fp32 (no scale constant — the
learned post_quant_conv is faithful)
- Qwen3 hidden states: cosine 1.0, max_abs 4.4e-4 (fp32 accumulation floor)
- 10-prompt perceptual A/B at 100 steps: passed human review (correct content,
duration, no tonal artifacts)
## Performance (Apple M5 Max)
100 steps, cfg 4.0, full 30 s latent: 60 s wall clock, 14.2 GB peak memory.
## License
Apache-2.0, matching the upstream model, code, and all components.
|