Enable CUTLASS grouped GEMM for pretraining wgrad on GB200 and H100 #5963
Triggered via pull request
September 17, 2025 03:09
jiawenliu64
Status
Success
Total duration
2h 27m 26s
Artifacts
5
build_wheels_genai_linux_x86.yml
on: pull_request
generate-matrix
/
generate
4s
Matrix: pytorch/FBGEMM / build
Matrix: pytorch/FBGEMM / upload / upload
Annotations
1 warning
|
filter-matrix
The `python-version` input is not set. The version of Python currently in `PATH` will be used.
|
Artifacts
Produced during runtime
| Name | Size | Digest | |
|---|---|---|---|
|
pytorch_FBGEMM__3.10_cu126_x86_64
|
17.5 MB |
sha256:1f1445aa0863f4c0a3204d47d87feda6b98a3f0a88c9e4aa451152d097aba62d
|
|
|
pytorch_FBGEMM__3.10_cu128_x86_64
|
49.3 MB |
sha256:06c66a3234a27bb23cf13116a65e35fbbe1fb245ecb0282c3c8a1f52a8c0cca3
|
|
|
pytorch_FBGEMM__3.10_cu130_x86_64
|
47.3 MB |
sha256:06633e79d1e86427804ce1182752022cd804a5c0363eba272f9997aa1bc6c80d
|
|
|
pytorch_FBGEMM__3.10_rocm6.3_x86_64
|
11.2 MB |
sha256:abbcd0db81f9966f8f69be5f7a0514660cf76151f6c402fe04690ea802e774bd
|
|
|
pytorch_FBGEMM__3.10_rocm6.4_x86_64
|
11.2 MB |
sha256:1e4d7bafe67d332ac2b7c41a10644f104416a513b559932e02e668b746120ca9
|
|