Skip to content
Back to pull request #18619

Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #21834

Sign in to view logs

Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM.

Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #21834

Triggered via pull request February 28, 2024 23:40
@chenfucnchenfucn
synchronize #18619
chenfucn:cfu_kernel
Status Success
Total duration 5m 19s
Artifacts –

codeql.yml

on: pull_request
Matrix: Analyze
Fit to window
Zoom out
Zoom in