---
license: cc-by-nc-4.0
language:
  - en
tags:
  - mlx
  - audio
  - music
  - audio-super-resolution
  - flashsr
  - latent-diffusion
  - quantized
  - int4
base_model: jakeoneijk/FlashSR_Inference
library_name: mlx
pipeline_tag: audio-to-audio
---

# FlashSR-MLX-4bit

- [speech-swift](https://github.com/soniqo/speech-swift) — Apple SDK
- [soniqo.audio](https://soniqo.audio) — website
- [blog](https://soniqo.audio/blog) — blog

MLX port of [FlashSR](https://github.com/jakeoneijk/FlashSR_Inference) — a
**single-step** distilled latent-diffusion audio super-resolution model —
quantized to **INT4** weight-only for on-device inference on Apple Silicon.
Upsamples any-rate input (mono) to **48 kHz** in one diffusion pass with a
specialised BigVGAN-flavour SR vocoder. FlashSR is the distilled student of
[AudioSR](https://github.com/haoheliu/versatile_audio_super_resolution).

## Model

| | |
|---|---|
| Total parameters | 638 M (VAE 223 M + UNet 258 M + Vocoder 157 M) |
| Diffusion steps | **1** (distilled v-prediction student, cosine schedule, t=999) |
| Quantization | INT4 weight-only, group size 64, mode `mlx_affine_flat` |
| Format | MLX safetensors (single combined bundle) |
| Sample rate | 48 kHz mono out (any-rate mono in) |
| Frame length | 5.12 s (245 760 samples) per forward |
| Bundle size | 346 MB on disk |
| Source | [jakeoneijk/FlashSR_Inference](https://github.com/jakeoneijk/FlashSR_Inference) |

## Files

| File | Size | Description |
|---|---|---|
| `model.safetensors` | 346 MB | INT4-quantized VAE + UNet + SR Vocoder weights |
| `config.json` | ~70 KB | Sub-model configs + quantization metadata + original shape table for dequant-on-load |

The three sub-models share one safetensors file with `vae.*`, `ldm.*`, `voc.*`
key prefixes. `config.quantized_shapes` records each tensor's pre-flatten shape
so `mx.dequantize` can rebuild conv weight tensors at load time.

## Performance (Apple Silicon, M-series, 5.12 s @ 48 kHz)

| Metric | Value |
|---|---|
| Real-time factor (wall / audio) | **1.10** |
| Load time | 0.17 s (dequant materialises bf16 weights once) |
| SNR vs FP16 reference | **+29.4 dB** |
| Cosine similarity vs FP16 | **0.9994** |
| Peak amplitude preservation | 1.000 |

INT4 is the recommended deployment variant — well above music-perceptual
threshold and the smallest viable on-device bundle.

## Usage

```python
from huggingface_hub import snapshot_download
import mlx.core as mx
import numpy as np
import scipy.io.wavfile as wf
from scipy.signal import resample_poly

bundle = snapshot_download("aufklarer/FlashSR-MLX-4bit")
# See https://github.com/soniqo/speech-swift for production usage.

# Toy Python demo (requires the matching MLX FlashSR runtime):
sr, audio = wf.read("lr.wav")
audio = audio.astype(np.float32) / 32767.0
audio_48 = resample_poly(audio, 48000, sr).astype(np.float32)

from flashsr import FlashSR  # from this repo's export/
model = FlashSR(bundle)
hr = model(mx.array(audio_48), seed=42)
mx.eval(hr)
wf.write("hr.wav", 48000, (np.clip(np.array(hr), -1, 1) * 32767).astype(np.int16))
```

## Source

- Upstream: [jakeoneijk/FlashSR_Inference](https://github.com/jakeoneijk/FlashSR_Inference)
- Paper: [FlashSR: One-step Versatile Audio Super-resolution via Diffusion Distillation](https://arxiv.org/abs/2501.10295)
- Teacher model: [AudioSR](https://github.com/haoheliu/versatile_audio_super_resolution)

## License

**CC-BY-NC 4.0** — inherited from upstream FlashSR weights. Non-commercial use only.