briaai/RMBG-2.0

Segmentation · served with ONNX Runtime · ↓ 391.6K downloads a month · View on Hugging Face

The model's publisher asks you to accept its licence on Hugging Face before using it. Accept it there with your Hugging Face account; GridPup will use your Hugging Face token to download it.

Weights: 1.0 GB
GPU memory needed: about 2.5 GB
Memory · disk: 8 GB · 30 GB
Licence: other

Versions

Smaller versions (8-bit, 4-bit) need less GPU memory and answer a little less precisely.

full precisiononnx/model.onnx1.0 GB · 2.5 GB GPU 4-bit (bnb)onnx/model_bnb4.onnx360 MB · 1.5 GB GPU float16onnx/model_fp16.onnx510 MB · 1.8 GB GPU 8-bitonnx/model_int8.onnx370 MB · 1.5 GB GPU 4-bitonnx/model_q4.onnx370 MB · 1.6 GB GPU 4-bit, float16onnx/model_q4f16.onnx230 MB · 1.4 GB GPU 8-bitonnx/model_quantized.onnx370 MB · 1.5 GB GPU 8-bitonnx/model_uint8.onnx370 MB · 1.5 GB GPU

Hardware and deploy

Deploying licence-gated models is coming soon: you will accept the licence on Hugging Face and give GridPup your Hugging Face token.

Loading…