Skip to content
Lint

Adding cuda kernel (optimized for sm80) for block-wise 4b quantized float 16 GEMM. #25753

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. #25753

Run time

Learn about OS pricing on GitHub Actions
Job Run time
Lint C++
22m 42s
Lint JavaScript
35s
Python format
4m 55s
Optional Lint
1m 3s
29m 15s