Hugging Face's logo

mlx-community
/
BiRefNet-fp16

Image Segmentation
MLX
Safetensors
BiRefNet
matting
background-removal
Model card Files
xet
 Community
BiRefNet-fp16 / README.md
xocialize's picture
xocialize
fp16 MLX conversion of ZhengPeng7/BiRefNet (Swin-L, 1024) — soft-alpha matting
b989bb7 verified
preview code
|
Raw
History Blame Contribute Delete
1.43 kB
metadata 
license: mit
base_model: ZhengPeng7/BiRefNet
library_name: mlx
tags:
  - mlx
  - image-segmentation
  - matting
  - background-removal
  - birefnet
pipeline_tag: image-segmentation

BiRefNet-fp16 (MLX)

mlx-community/BiRefNet-fp16 is an fp16 MLX conversion of ZhengPeng7/BiRefNet (MIT) — a Swin-L + ASPP-Deformable foreground segmentation / matting model at 1024×1024, producing a single-channel soft-alpha matte (white = foreground). The fast, general-purpose tier.

Parity: IoU 0.9905 vs the PyTorch reference (zero unused keys). fp16 runtime validated for production matting quality.

Use (Swift / MLX)

Loaded by mlx-birefnet-swift — the vendored BiRefNet core plus a conformant MLXEngine matting ModelPackage:

import BiRefNet
let pipeline = try BiRefNetPipeline.fromPretrained("model.safetensors", dtype: .float16)  // inputSize 1024
let matte = try pipeline(cgImage).maskCGImage()   // source-resolution soft-alpha

Converted from the official PyTorch checkpoint via the package's birefnet-convert (PyTorch NCHW → MLX NHWC; 754 → 687 tensors). Single-file model.safetensors. See also the higher-res tier mlx-community/BiRefNet_HR-matting-fp16.