Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #18619
Azure Pipelines / Android CI Pipeline (BUILD_CPU_STAGE Build_CPU_EP)
succeeded
Feb 29, 2024 in 19m 38s
BUILD_CPU_STAGE Build_CPU_EP succeeded
Loading