This repository was archived by the owner on Mar 30, 2026. It is now read-only.
Repository navigation
This repository was archived by the owner on Mar 30, 2026. It is now read-only.
[question] Do we have a method for 2-bit quantization? #872
Description
Describe the issue
like title: Do we have a method for 2-bit quantization on intel iGPU?
Or Can our hardware machine support 2-bit quantization? Are there any other open-source methods（openvino,pytorch） that have implemented 2-bit quantitative inference on igpu.
my igpu is :
Intel(R) Arc(TM) 140V GPU (16GB)
or:
Intel(R) Arc(TM) 140T GPU (16GB)
Activity