Migrate GenAI gqa attn splitk kernels to FBGEMM_LAUNCH_KERNEL
, pt 2…
#15421
fbgemm_gpu_ci_cuda.yml
on: push
generate-build-matrix
/
generate_ci_matrix
7s
Matrix: build / build_artifact
Matrix: test / test_and_publish_artifact
Artifacts
Produced during runtime
Name | Size | Digest | |
---|---|---|---|
fbgemm_default_x86_gcc_py3.13_cu12.9.1.whl
|
899 MB |
sha256:f7ae417d2dde09097cd99e349ca0273c506c569784a132f7126fcaf10eb62db0
|
|
fbgemm_genai_x86_gcc_py3.13_cu12.9.1.whl
|
49.6 MB |
sha256:b56ad3340085f7df479c440e359bf6b3589aabc035392abe225b6ed6ad008a8f
|
|
fbgemm_hstu_x86_gcc_py3.13_cu12.9.1.whl
|
21.8 MB |
sha256:85f809ed5c888b4260bf3624b63e3c413d5a2c134722ff560e52af4e8a2ae135
|
|