mlboydaisuke / MODNet-ExecuTorch

huggingface.co
Total runs: 90
24-hour runs: 0
7-day runs: 12
30-day runs: -46
Model's Last Updated: August 15 2026
image-segmentation

Introduction of MODNet-ExecuTorch

Model Details of MODNet-ExecuTorch

modnet_portrait_matting — ExecuTorch XNNPACK

  • Source : ZHKKKe/MODNet + DavG25/modnet-pretrained-models ckpt
  • License : Apache-2.0
  • Input : [[1, 3, 512, 512]] — RGB [-1,1], 512x512
  • Output : alpha matte [1,1,512,512] 0-1
Variants

All variants take and return fp32 tensors — swap the .pte file, keep your app code.

precision file size (MB) parity vs fp32 eager (worst corr) Mac median (ms)*
fp32 modnet_portrait_matting_xnnpack_fp32.pte 26.1 1.000000 64.9
fp16 modnet_portrait_matting_xnnpack_fp16.pte 24.4 1.000000 107.8
int8 modnet_portrait_matting_xnnpack_int8.pte 6.8 0.999949 46.5

*Mac arm64, single process, median of 10 — a reference point for relative cost only, not a device number (torch eager fp32 on the same machine: 120.1 ms).

Checked in the task's own units

Correlation is a first filter. These are the numbers that decide:

  • int8 — measured in the units that matter for this model: mask IoU at 0.5, median 0.9922 over five real images (worst 0.9613) against the fp32 build.
Verification (executorch 1.4.0, torch 2.13.0)

Parity is measured against the fp32 eager model on real image input; corr is the correlation over all elements of each output tensor.

output shape max_abs_diff corr
0 [1, 1, 512, 512] 1.476e-04 1.000000

XNNPACK delegate coverage (fp32): 94.7% (302/319 ops); ops left on the portable kernels: aten._native_batch_norm_legit.no_stats x16, aten.expand_copy.default x1

Conversion

torch.export -> to_edge_transform_and_lower(XnnpackPartitioner) -> .pte (conversion scripts: executorch-models )

Runs of mlboydaisuke MODNet-ExecuTorch on huggingface.co

90
Total runs
0
24-hour runs
-2
3-day runs
12
7-day runs
-46
30-day runs

More Information About MODNet-ExecuTorch huggingface.co Model

More MODNet-ExecuTorch license Visit here:

https://choosealicense.com/licenses/apache-2.0

MODNet-ExecuTorch huggingface.co

MODNet-ExecuTorch huggingface.co is an AI model on huggingface.co that provides MODNet-ExecuTorch's model effect (), which can be used instantly with this mlboydaisuke MODNet-ExecuTorch model. huggingface.co supports a free trial of the MODNet-ExecuTorch model, and also provides paid use of the MODNet-ExecuTorch. Support call MODNet-ExecuTorch model through api, including Node.js, Python, http.

mlboydaisuke MODNet-ExecuTorch online free

MODNet-ExecuTorch huggingface.co is an online trial and call api platform, which integrates MODNet-ExecuTorch's modeling effects, including api services, and provides a free online trial of MODNet-ExecuTorch, you can try MODNet-ExecuTorch online for free by clicking the link below.

mlboydaisuke MODNet-ExecuTorch online free url in huggingface.co:

https://huggingface.co/mlboydaisuke/MODNet-ExecuTorch

MODNet-ExecuTorch install

MODNet-ExecuTorch is an open source model from GitHub that offers a free installation service, and any user can find MODNet-ExecuTorch on GitHub to install. At the same time, huggingface.co provides the effect of MODNet-ExecuTorch install, users can directly use MODNet-ExecuTorch installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

MODNet-ExecuTorch install url in huggingface.co:

https://huggingface.co/mlboydaisuke/MODNet-ExecuTorch

Url of MODNet-ExecuTorch

MODNet-ExecuTorch huggingface.co Url

Provider of MODNet-ExecuTorch huggingface.co

mlboydaisuke
ORGANIZATIONS

Other API from mlboydaisuke