SAVRN
Search Contact SAVRN

Open-weight model · Image segmentation

RMBG-2.0

by BRIA AI briaai/RMBG-2.0

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"…

Parameters221M
Context—
Weights5.4 GB
Licenseother
AccessAccess requested at publisher
Monthly Downloads639.3k

Runs On

What it takes to serve RMBG-2.0 (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 Sep 18, 2026.

SAVRN's Notes on RMBG-2.0

Half a gigabyte of memory at 16-bit, with 0.4 GB of weights, means the hardware question for RMBG-2.0 is not which GPU can hold it but what else runs beside it. The cheapest on-demand card on our board, one MI300X with 192 GB at $1.85 an hour, would sit almost entirely idle under a 221M-parameter BiRefNet segmentation model, so we treat it as a stage inside a larger image pipeline rather than a workload that earns its own machine. BRIA trained it on stock, e-commerce, gaming and advertising imagery.

The license reads other, not a standard permissive grant, and access is gated, so a commercial deployment starts with a request to BRIA and their terms in hand. Also check disk: the repository is 19 files and 5.4 GB across safetensors, ONNX and PyTorch, well beyond the weights you load, last updated April 6, 2026.

SAVRN Research, 2026-09-18

Model Card

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…

Excerpt from the card by BRIA AI, licensed other.

Identity and Version

Repository
briaai/RMBG-2.0
Publisher
BRIA AI
Task
Image segmentation
Modality
Image
Library
transformers
Parameters
221M parameters
Languages
Not stated by the source
Revision
5df4c9c76d8170882c34f6986e848ee07fd0ba43
First published
2024-10-29
Last updated
2026-04-06

Files and Weights

19 files, 5.4 GB in total. The weights are 10 files totalling 5.4 GB in bin, onnx, safetensors.

Weights10 files · 5.4 GB
Configuration4 files · 92.5 KB
Documentation1 file · 9.4 KB
Other3 files · 6.7 MB
Repository1 file · 1.8 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights884.9 MB —
onnx/model.onnxWeights1.0 GB —
onnx/model_bnb4.onnxWeights355.3 MB —
onnx/model_fp16.onnxWeights513.6 MB —
onnx/model_int8.onnxWeights366.1 MB —
onnx/model_q4.onnxWeights367.5 MB —
onnx/model_q4f16.onnxWeights233.8 MB —
onnx/model_quantized.onnxWeights366.1 MB —
onnx/model_uint8.onnxWeights366.1 MB —
pytorch_model.binWeights885.1 MB —
BiRefNet_config.pyConfiguration298 B —
birefnet.pyConfiguration91.4 KB —
config.jsonConfiguration405 B —
preprocessor_config.jsonConfiguration391 B —
README.mdDocumentation9.4 KB —
collage5.pngOther4.5 MB —
diagram1.pngOther21.3 KB —
t4.pngOther2.2 MB —
.gitattributesRepository1.8 KB —

License and Download

License
other
Access
Access requested at publisher
Download size
5.4 GB
Request access from BRIA AI

BRIA AI grants access through its official repository on Hugging Face. Read the license.

Memory Requirements

PrecisionWeights in memory
As published5.4 GB
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 RMBG-2.0

How much GPU memory does RMBG-2.0 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 RMBG-2.0 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.

What license is RMBG-2.0 released under?

other, as its publisher declares it. Read the license text before commercial use.

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

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_HR-matting

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