Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Deploy nightly docs

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
deploy_nightly_docs.yml
889 workflow runs
889 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
[Common] Ensure quantization kernels handle noop properly (#3271) Deploy nightly docs #1961: Commit 172bd93 pushed by ptrendx
1m 54s main
main
1m 54s
[Pytorch] Optimize CPU overhead for NCCL EP eager mode (#3378) Deploy nightly docs #1960: Commit e61a2fb pushed by phu0ngng
2m 16s main
main
2m 16s
Fix FlashAttention local version validation (#3356) Deploy nightly docs #1959: Commit a0a2d08 pushed by vcherepanov-nv
1m 48s main
main
1m 48s
[JAX] Move ffi type definition before ffi target definitions (#3382) Deploy nightly docs #1958: Commit 86e3487 pushed by jberchtold-nvidia
2m 0s main
main
2m 0s
[PyTorch] Enable NVFP4 row-scaled (per-token) backward for GroupedLin… Deploy nightly docs #1957: Commit 621da35 pushed by timmoon10
1m 51s main
main
1m 51s
Add 2d quant for mxfp8 (#2634) Deploy nightly docs #1956: Commit d8815c4 pushed by ptrendx
2m 54s main
main
2m 54s
[Pytorch] Enable TE Op to consume extra_outputs from a previously run… Deploy nightly docs #1955: Commit a22f4cd pushed by vthumbe1503
2m 36s main
main
2m 36s
fix: typo and unnecessary f-string in BRCM skip message (#3370) Deploy nightly docs #1954: Commit c280e7c pushed by cyanguwa
7m 0s main
main
7m 0s
Stabilize FlashAttention CI environments (#3341) Deploy nightly docs #1953: Commit 7d0365b pushed by sudhakarsingh27
5m 28s main
main
5m 28s
fix: remove unused import time (#3364) Deploy nightly docs #1952: Commit 900ac82 pushed by cyanguwa
6m 53s main
main
6m 53s
fix: remove redundant second export of XLA_FLAGS in L0 (#3377) Deploy nightly docs #1951: Commit 910be8b pushed by cyanguwa
3m 2s main
main
3m 2s
fix: allgather striped window_size (5,0) id says (8,0) (#3374) Deploy nightly docs #1950: Commit 67c26e2 pushed by cyanguwa
1m 48s main
main
1m 48s
Improve device-init grouped linear module with single grouped weight … Deploy nightly docs #1949: Commit 2d80391 pushed by vthumbe1503
2m 3s main
main
2m 3s
[Common] Upgrade nccl-extensions commit (#3349) Deploy nightly docs #1948: Commit 779f17d pushed by phu0ngng
2m 9s main
main
2m 9s
[Pytorch] Fix swizzle scale output shape in variable shape case (#3344) Deploy nightly docs #1947: Commit 1ad30c6 pushed by YangFei1990
2m 18s main
main
2m 18s
[Common] Fix NVFP4 stochastic rounding on architectures without cvt.r… Deploy nightly docs #1946: Commit fc1fc0e pushed by ptrendx
1m 51s main
main
1m 51s
[PyTorch] Advance FusedAdam step counter for empty param groups (#3318) Deploy nightly docs #1945: Commit 734ce31 pushed by ptrendx
2m 2s main
main
2m 2s
[Common][PyTorch] EP dispatch with unfused MXFP8 quantization (#3270) Deploy nightly docs #1944: Commit 56a05e7 pushed by phu0ngng
1m 56s main
main
1m 56s
Enable runtime resolution of CUDA header path for NVRTC (#3252) Deploy nightly docs #1943: Commit b54bb1e pushed by ptrendx
2m 1s main
main
2m 1s
[Common] MXFP8: Add swizzled-scale fast path for cast-only quantizati… Deploy nightly docs #1942: Commit 07e281f pushed by Oleg-Goncharov
2m 2s main
main
2m 2s
[Common] Fix pointer arithmatic to generate correct LDS/STS instructi… Deploy nightly docs #1941: Commit f07a860 pushed by ptrendx
2m 1s main
main
2m 1s
[Bug] use_single_group_runtime_offsets arg is not supported for Scale… Deploy nightly docs #1940: Commit 598b9ea pushed by vthumbe1503
1m 58s main
main
1m 58s
[JAX][Fix] Gate collective-stream compute_on on import and silence py… Deploy nightly docs #1939: Commit 32f7096 pushed by phu0ngng
2m 0s main
main
2m 0s
[PyTorch] Preserve FP8 recompute state for inner autocast (#3284) Deploy nightly docs #1938: Commit 731f242 pushed by pggPL
2m 4s main
main
2m 4s
Use a temp directory for CMake build (#3254) Deploy nightly docs #1937: Commit 8260f49 pushed by ptrendx
2m 1s main
main
2m 1s