Fix handling of dynamic FP8 grouped gemm on Nvidia #8985
jwfrommbuild_wheels_linux_aarch64.yml
on: pull_request
generate-matrix
/
generate
7s
Matrix: build
Artifacts
Produced during runtime
| Name | Size | |
|---|---|---|
|
pytorch_FBGEMM__3.9_cpu_aarch64
|
2.7 MB |
|
|
pytorch_FBGEMM__3.9_cu126_aarch64
|
446 MB |
|