Skip to content

Optimize fused gemm+dequant kernel for ROCm, use it for batch sizes o…

6aeade4
Select commit
Loading
Failed to load commit list.
Open

[ROCm] Optimize kgemm_4bit_inference_naive for ROCm, use it for batch sizes other than 1 #1920

Optimize fused gemm+dequant kernel for ROCm, use it for batch sizes o…
6aeade4
Select commit
Loading
Failed to load commit list.