Skip to page content
SAVRN
Search Plan a project

Open-weight model · Image segmentation

BiRefNet_HR-matting

by Peng Zheng ZhengPeng7/BiRefNet_HR-matting

BiRefNet_HR-matting is an open-weight model for image segmentation from Peng Zheng, released under MIT License. It has 221M parameters. At 16-bit it needs about 0.5 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 41.4k downloads a month.

This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024). Visit our GitHub repo: https://github.com/ZhengPeng7/BiRefNet for more details -- codes, docs, and model zoo!

Parameters221M
Context—
Weights444.5 MB
Licensemit
AccessOpen weights
Monthly Downloads41.4k

Runs On

What it takes to serve BiRefNet_HR-matting (221M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 0.4 GB 0.5 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.2 GB 0.3 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.1 GB 0.1 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 16, 2026.

BiRefNet_HR-matting on every accelerator the SAVRN Index prices, at every precision

Model Card

By Peng Zheng, published under mit, revision 5d6b6f8adcb5.

This BiRefNet was trained with images in 2048x2048 for higher resolution image matting with transparency.

Performance:

All tested in FP16 mode.

Dataset Method Resolution maxFm wFmeasure MAE Smeasure meanEm HCE maxEm meanFm adpEm adpFm mBA maxBIoU meanBIoU
TE-AM-2k BiRefNet_HR-matting-epoch_135 2048x2048 .974 .997 .989 .002 .998 .987 .988 .961 .981 .000 .879 .965 .893
TE-P3M-500-NP BiRefNet_HR-matting-epoch_135 2048x2048 .980 .996 .989 .002 .997 .987 .989 .880 .900 .000 .853 .947 .897
TE-AM-2k BiRefNet-matting-epoch_100 1024x1024 .973 .996 .990 .003 .997 .987 .989 .987 .991 .000 .846 .952 .890
TE-P3M-500-NP BiRefNet-matting-epoch_100 1024x1024 .979 .996 .990 .003 .997 .987 .989 .928 .951 .000 .830 .940 .891
TE-AM-2k BiRefNet-matting-epoch_100 2048x2048 .971 .996 .990 .003 .997 .987 .988 .990 .992 .000 .838 .941 .891
TE-P3M-500-NP BiRefNet-matting-epoch_100 2048x2048 .978 .995 .990 .003 .996 .987 .989 .955 .971 .000 .818 .931 .891

Bilateral Reference for High-Resolution Dichotomous Image Segmentation

Read the full model card (785 words)

Configuration

Architecture
BiRefNet

Identity and Version

Repository
ZhengPeng7/BiRefNet_HR-matting
Publisher
Peng Zheng
Task
Image segmentation
Modality
Image
Library
birefnet
Parameters
221M parameters
Languages
Not stated by the source
Revision
5d6b6f8adcb5b417c871b1d84ceaae9871355b7f
First published
2025-02-12
Last updated
2026-02-04

Files and Weights

8 files, 444.6 MB in total. The weights are 1 file totalling 444.5 MB in safetensors.

Weights1 file · 444.5 MB
Configuration4 files · 97.6 KB
Documentation1 file · 11.5 KB
Other1 file · 149 B
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights444.5 MB a5a4de698739
BiRefNet_config.pyConfiguration298 B —
birefnet.pyConfiguration92.1 KB —
config.jsonConfiguration416 B —
handler.pyConfiguration4.8 KB —
README.mdDocumentation11.5 KB —
requirements.txtOther149 B —
.gitattributesRepository1.5 KB —

License and Download

License
mit
Access
Open weights, no gate
Download size
444.5 MB
Download from Peng Zheng

Released by Peng Zheng through its official repository on Hugging Face. Read the license.

Built From

Memory Requirements

PrecisionWeights in memory
As published444.5 MB
16-bit0.4 GB
8-bit0.2 GB
4-bit0.1 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About BiRefNet_HR-matting

How much GPU memory does BiRefNet_HR-matting need?

About 0.5 GB at 16-bit and 0.1 GB at 4-bit: the weights (221M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run BiRefNet_HR-matting on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use BiRefNet_HR-matting commercially?

Yes. BiRefNet_HR-matting is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.

Similar Models

Model · Image segmentation

BiRefNet

Peng Zheng

This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024). Visit our GitHub repo: https://github.com/ZhengPeng7/BiRefNet for more details -- codes, docs, and model zoo! This repo contains the weights of BiRefNet proposed in our paper, which has achieved the SOTA performance on three tasks (DIS, HRSOD, and COD). Go to my GitHub page for BiRefNet codes and the latest updates: https://github.com/ZhengPeng7/BiRefNet:) + Online Image Inference on Colab: + Online Inference with GUI on Hugging Face with adjustable resolutions: + Inference and evaluation of your given weights: + Many thanks to @Freepik for their generous…

Open weights mit 221M parameters birefnet
View model

Model · Image segmentation

RMBG-2.0

BRIA AI

href="https://huggingface.co/briaai/FIBO" target="blank" rel="noopener" aria-label="Explore FIBO on Hugging Face" style=" src="https://huggingface.co/front/assets/huggingfacelogo-noborder.svg" alt="Hugging Face" width="18" height="18" style="display:block" RMBG v2.0 is our new state-of-the-art background removal model significantly improves RMBG v1.4. The model is designed to effectively separate foreground from background in a range of categories and image types. This model has been trained on a carefully selected dataset, which includes: general stock images, e-commerce, gaming, and advertising content, making it suitable for commercial use cases powering enterprise content creation at…

Access requested at publisher other 221M parameters transformers
View model

Model · Image segmentation

BiRefNet_HR

Peng Zheng

This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024). Visit our GitHub repo: https://github.com/ZhengPeng7/BiRefNet for more details -- codes, docs, and model zoo! This repo contains the weights of BiRefNet proposed in our paper, which has achieved the SOTA performance on three tasks (DIS, HRSOD, and COD). Go to my GitHub page for BiRefNet codes and the latest updates: https://github.com/ZhengPeng7/BiRefNet:) + Online Image Inference on Colab: + Online Inference with GUI on Hugging Face with adjustable resolutions: + Inference and evaluation of your given weights: + Many thanks to @freepik for their generous…

Open weights mit 221M parameters birefnet
View model

Model · Image segmentation

BiRefNet-portrait

Peng Zheng

Check the main BiRefNet model repo for more info and how to use it: https://huggingface.co/ZhengPeng7/BiRefNet/blob/main/README.md Also check the GitHub repo of BiRefNet for all things you may want: https://github.com/ZhengPeng7/BiRefNet + Many thanks to @fal for their generous support on GPU resources for training this BiRefNet for portrait matting.

Open weights mit 221M parameters birefnet
View model

Model · Image segmentation

BiRefNet_dynamic

Peng Zheng

For performance of different epochs, check the evalresults-xxx folder for it on my google drive. This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024). Visit our GitHub repo: https://github.com/ZhengPeng7/BiRefNet for more details -- codes, docs, and model zoo! This repo contains the weights of BiRefNet proposed in our paper, which has achieved the SOTA performance on three tasks (DIS, HRSOD, and COD). Go to my GitHub page for BiRefNet codes and the latest updates: https://github.com/ZhengPeng7/BiRefNet:) + Online Image Inference on Colab: + Online Inference with GUI on Hugging Face with adjustable resolutions: +…

Open weights mit 221M parameters birefnet
View model

Model · Image segmentation

BiRefNet-matting

Peng Zheng

Check the main BiRefNet model repo for more info and how to use it: https://huggingface.co/ZhengPeng7/BiRefNet/blob/main/README.md Also check the GitHub repo of BiRefNet for all things you may want: https://github.com/ZhengPeng7/BiRefNet + Many thanks to @freepik for their generous support on GPU resources for training this model!

Open weights mit 221M parameters birefnet
View model