Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Deploy nightly docs

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
deploy_nightly_docs.yml
923 workflow runs
923 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Thread tanh logit softcapping through FlashAttention (FA2 and FA3) (#… Deploy nightly docs #2057: Commit f72bc1a pushed by cyanguwa
1m 50s main
main
1m 50s
[JAX] Interpolate the tensor-sequence-parallelism warning (#3486) Deploy nightly docs #2056: Commit 02962a9 pushed by jberchtold-nvidia
1m 51s main
main
1m 51s
[All] Refactor fused attention APIs with cuDNN-frontend support check… Deploy nightly docs #2055: Commit f689f8f pushed by cyanguwa
1m 54s main
main
1m 54s
Skip fetching unused next tile in Mxfp8 quantization kernel (#3500) Deploy nightly docs #2054: Commit 1634a5a pushed by ptrendx
1m 45s main
main
1m 45s
[JAX] Hopper BF16 grouped GEMM v2 support (#3083) Deploy nightly docs #2053: Commit 65809a1 pushed by jberchtold-nvidia
1m 44s main
main
1m 44s
[Common] Fix Grouped MXFP8 work mapping and TMA synchronization (#3483) Deploy nightly docs #2052: Commit 7ba2f9f pushed by Oleg-Goncharov
1m 33s main
main
1m 33s
[PyTorch] Fix NaN expert weight gradients at num_groups == 1 with SRe… Deploy nightly docs #2051: Commit 39a5401 pushed by vthumbe1503
1m 57s main
main
1m 57s
Add entrypoint for flagos multi-backend plugin system (#3401) Deploy nightly docs #2050: Commit 7c6fa59 pushed by ptrendx
1m 47s main
main
1m 47s
[PyTorch] Use torch's register_custom_class API for opaque quantizers… Deploy nightly docs #2049: Commit b97f5ab pushed by pggPL
1m 47s main
main
1m 47s
Fix hybrid fp8 test guard (#3492) Deploy nightly docs #2048: Commit cd24504 pushed by negvet
1m 40s main
main
1m 40s
[PyTorch] Document fine-grained quantization recipes (#3336) Deploy nightly docs #2047: Commit f2bec23 pushed by pggPL
1m 51s main
main
1m 51s
Update Linear docstring referring torch.Linear (#3491) Deploy nightly docs #2046: Commit dc8909d pushed by negvet
1m 28s main
main
1m 28s
[Common] Preserve shared-memory pointer provenance in TMA kernels (#3… Deploy nightly docs #2045: Commit 5f6105b pushed by ptrendx
1m 38s main
main
1m 38s
fix: Unreachable backend check after earlier backend skip (#3368) Deploy nightly docs #2044: Commit a30aee5 pushed by ptrendx
2m 16s main
main
2m 16s
Relax runtime checks for activation recompute into Warnings (#3436) Deploy nightly docs #2043: Commit 846379d pushed by timmoon10
1m 56s main
main
1m 56s
[PyTorch] Reduce CUDA graph memory retention (#3427) Deploy nightly docs #2042: Commit 29229b5 pushed by timmoon10
1m 24s main
main
1m 24s
Use cuDNN's deterministic dprob in the fused grouped MLP (#3407) Deploy nightly docs #2041: Commit d1e9c33 pushed by vthumbe1503
2m 27s main
main
2m 27s
Reduce Grouped MLP Fuser CPU Overhead (#3410) Deploy nightly docs #2040: Commit 80a89ad pushed by timmoon10
1m 59s main
main
1m 59s
[NCCL][PyTorch] Fix test-fusible-ops-file-rendezvous-bug (#3478) Deploy nightly docs #2039: Commit 1f25d6a pushed by timmoon10
2m 35s main
main
2m 35s
[PyTorch] Fix: Resolve EP symm-mem window offset for both old and new… Deploy nightly docs #2038: Commit e61b66d pushed by phu0ngng
1m 40s main
main
1m 40s
[PyTorch] Decline fused grouped MLP when the backward format is not E… Deploy nightly docs #2037: Commit b86d8f9 pushed by ptrendx
1m 47s main
main
1m 47s
[PyT] Fix lint failures (#3476) Deploy nightly docs #2036: Commit 45808f1 pushed by cyanguwa
2m 10s main
main
2m 10s
WAR for the cuBLAS grouped gemm issue (#3475) Deploy nightly docs #2035: Commit ee253aa pushed by ptrendx
2m 6s main
main
2m 6s
[JAX] Add sqrtsoftplus router score function (#3448) Deploy nightly docs #2034: Commit 53d8981 pushed by jberchtold-nvidia
1m 41s main
main
1m 41s
fix: make_mask uses deprecated jnp.bool instead of jnp.bool_ (#3371) Deploy nightly docs #2033: Commit 362d77a pushed by ptrendx
2m 54s main
main
2m 54s