Skip to content

Allow Int4WeightOnlyQuantizer to set different dtype for scales_and_zeros #2030

Allow Int4WeightOnlyQuantizer to set different dtype for scales_and_zeros

Allow Int4WeightOnlyQuantizer to set different dtype for scales_and_zeros #2030

test (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://download.pytorc...  /  linux-job

succeeded Jul 5, 2024 in 31m 46s