Skip to content

Actions: tdophung/TransformerEngine

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
17 workflow runs
17 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[Common] Disabled the tuned NVFP4 kernels (#2615) Deploy nightly docs #13: Commit a0a89a8 pushed by tdophung
1m 20s main
main
1m 20s
[JAX] Fix cb.CUDAOptions usage for Triton 3.6.0 (#2610) Deploy nightly docs #12: Commit 8bf37f0 pushed by tdophung
1m 14s main
main
1m 14s
Fixed the year to 2026 (#2611) Deploy nightly docs #11: Commit 36f4e45 pushed by tdophung
1m 25s main
main
1m 25s
[PyTorch] Bunch of fixes for cpu offloading (#2535) Deploy nightly docs #10: Commit fe8fad5 pushed by tdophung
2m 9s main
main
2m 9s
[docs] Getting started refactor (#2534) Deploy nightly docs #9: Commit a976740 pushed by tdophung
1m 31s main
main
1m 31s
Document environment variables (#2552) Deploy nightly docs #8: Commit 27dc83b pushed by tdophung
1m 46s main
main
1m 46s
Fix ptxas compilation on sm103 for triton kernels (#2539) Deploy nightly docs #7: Commit 97a09c2 pushed by tdophung
1m 28s main
main
1m 28s
Jax primitives for permutation on single GPU (#2473) Deploy nightly docs #6: Commit 46c6ef3 pushed by tdophung
1m 30s main
main
1m 30s
[PyTorch] Change order of args in another permutation triton kernel … Deploy nightly docs #5: Commit e05f87e pushed by tdophung
1m 37s main
main
1m 37s
[Pytorch][Bug]MXFP8 Split tensor Bug fix (#2427) Deploy nightly docs #4: Commit c09411d pushed by tdophung
1m 25s main
main
1m 25s
Add primary weighs fp8 support for mxfp8 (#2055) Deploy nightly docs #3: Commit d126cdd pushed by tdophung
1m 30s main
main
1m 30s
Set 0 as the size of dummy tensors to reduce memory usage. Documentation #1: Pull request #1 opened by mingxu1067
1m 55s mingxu1067:mingh/set_dummy_with_zero_sizes
mingxu1067:mingh/set_dummy_with_zero_sizes
1m 55s
Set 0 as the size of dummy tensors to reduce memory usage. Lint #1: Pull request #1 opened by mingxu1067
3m 16s mingxu1067:mingh/set_dummy_with_zero_sizes
mingxu1067:mingh/set_dummy_with_zero_sizes
3m 16s
Set 0 as the size of dummy tensors to reduce memory usage. Build #1: Pull request #1 opened by mingxu1067
1h 52m 20s mingxu1067:mingh/set_dummy_with_zero_sizes
mingxu1067:mingh/set_dummy_with_zero_sizes
1h 52m 20s
Set 0 as the size of dummy tensors to reduce memory usage. License #1: Pull request #1 opened by mingxu1067
11s mingxu1067:mingh/set_dummy_with_zero_sizes
mingxu1067:mingh/set_dummy_with_zero_sizes
11s
Docs fix (#2301) Deploy nightly docs #2: Commit df39a7c pushed by tdophung
1m 41s main
main
1m 41s
[PyTorch] Only disable Flash Attention in Userbuffers test on SM 8.0 … Deploy nightly docs #1: Commit f8cb598 pushed by tdophung
1m 47s main
main
1m 47s