Deploy nightly docs

Workflow runs for Deploy nightly docs

• Skip fetching unused next tile in Mxfp8 quantization kernel (#3500)

  completed successfully:
  Deploy nightly docs#2054:Commit1634a5apushedbyptrendx
  main

  Sep 10, 20261m 45s
  More actions for workflow run 2054

• [JAX] Hopper BF16 grouped GEMM v2 support (#3083)

  completed successfully:
  Deploy nightly docs#2053:Commit65809a1pushedbyjberchtold-nvidia
  main

  Sep 9, 20261m 44s
  More actions for workflow run 2053

• [Common] Fix Grouped MXFP8 work mapping and TMA synchronization (#3483)

  completed successfully:
  Deploy nightly docs#2052:Commit7ba2f9fpushedbyOleg-Goncharov
  main

  Sep 9, 20261m 33s
  More actions for workflow run 2052

• [PyTorch] Fix NaN expert weight gradients at num_groups == 1 with SRe…

  completed successfully:
  Deploy nightly docs#2051:Commit39a5401pushedbyvthumbe1503
  main

  Sep 9, 20261m 57s
  More actions for workflow run 2051

• Add entrypoint for flagos multi-backend plugin system (#3401)

  completed successfully:
  Deploy nightly docs#2050:Commit7c6fa59pushedbyptrendx
  main

  Sep 9, 20261m 47s
  More actions for workflow run 2050

• [PyTorch] Use torch's register_custom_class API for opaque quantizers…

  completed successfully:
  Deploy nightly docs#2049:Commitb97f5abpushedbypggPL
  main

  Sep 8, 20261m 47s
  More actions for workflow run 2049

• Fix hybrid fp8 test guard (#3492)

  completed successfully:
  Deploy nightly docs#2048:Commitcd24504pushedbynegvet
  main

  Sep 8, 20261m 40s
  More actions for workflow run 2048

• [PyTorch] Document fine-grained quantization recipes (#3336)

  completed successfully:
  Deploy nightly docs#2047:Commitf2bec23pushedbypggPL
  main

  Sep 8, 20261m 51s
  More actions for workflow run 2047

• Update Linear docstring referring torch.Linear (#3491)

  completed successfully:
  Deploy nightly docs#2046:Commitdc8909dpushedbynegvet
  main

  Sep 8, 20261m 28s
  More actions for workflow run 2046

• [Common] Preserve shared-memory pointer provenance in TMA kernels (#3…

  completed successfully:
  Deploy nightly docs#2045:Commit5f6105bpushedbyptrendx
  main

  Sep 5, 20261m 38s
  More actions for workflow run 2045

• fix: Unreachable backend check after earlier backend skip (#3368)

  completed successfully:
  Deploy nightly docs#2044:Commita30aee5pushedbyptrendx
  main

  Sep 5, 20262m 16s
  More actions for workflow run 2044

• Relax runtime checks for activation recompute into Warnings (#3436)

  completed successfully:
  Deploy nightly docs#2043:Commit846379dpushedbytimmoon10
  main

  Sep 4, 20261m 56s
  More actions for workflow run 2043

• [PyTorch] Reduce CUDA graph memory retention (#3427)

  cancelled:
  Deploy nightly docs#2042:Commit29229b5pushedbytimmoon10
  main

  Sep 4, 20261m 24s
  More actions for workflow run 2042

• Use cuDNN's deterministic dprob in the fused grouped MLP (#3407)

  completed successfully:
  Deploy nightly docs#2041:Commitd1e9c33pushedbyvthumbe1503
  main

  Sep 4, 20262m 27s
  More actions for workflow run 2041

• Reduce Grouped MLP Fuser CPU Overhead (#3410)

  completed successfully:
  Deploy nightly docs#2040:Commit80a89adpushedbytimmoon10
  main

  Sep 4, 20261m 59s
  More actions for workflow run 2040

• [NCCL][PyTorch] Fix test-fusible-ops-file-rendezvous-bug (#3478)

  completed successfully:
  Deploy nightly docs#2039:Commit1f25d6apushedbytimmoon10
  main

  Sep 4, 20262m 35s
  More actions for workflow run 2039

• [PyTorch] Fix: Resolve EP symm-mem window offset for both old and new…

  completed successfully:
  Deploy nightly docs#2038:Commite61b66dpushedbyphu0ngng
  main

  Sep 4, 20261m 40s
  More actions for workflow run 2038

• [PyTorch] Decline fused grouped MLP when the backward format is not E…

  completed successfully:
  Deploy nightly docs#2037:Commitb86d8f9pushedbyptrendx
  main

  Sep 4, 20261m 47s
  More actions for workflow run 2037

• [PyT] Fix lint failures (#3476)

  completed successfully:
  Deploy nightly docs#2036:Commit45808f1pushedbycyanguwa
  main

  Sep 4, 20262m 10s
  More actions for workflow run 2036

• WAR for the cuBLAS grouped gemm issue (#3475)

  completed successfully:
  Deploy nightly docs#2035:Commitee253aapushedbyptrendx
  main

  Sep 4, 20262m 6s
  More actions for workflow run 2035

• [JAX] Add sqrtsoftplus router score function (#3448)

  completed successfully:
  Deploy nightly docs#2034:Commit53d8981pushedbyjberchtold-nvidia
  main

  Sep 3, 20261m 41s
  More actions for workflow run 2034

• fix: make_mask uses deprecated jnp.bool instead of jnp.bool_ (#3371)

  completed successfully:
  Deploy nightly docs#2033:Commit362d77apushedbyptrendx
  main

  Sep 3, 20262m 54s
  More actions for workflow run 2033

• Compress CUDA binaries (#3465)

  completed successfully:
  Deploy nightly docs#2032:Commit888da30pushedbyptrendx
  main

  Sep 3, 20261m 59s
  More actions for workflow run 2032

• [PyTorch] Pair delayed-scaling FP8 recompute metadata per module (#3394)

  completed successfully:
  Deploy nightly docs#2031:Commitd16c754pushedbypggPL
  main

  Sep 3, 20262m 16s
  More actions for workflow run 2031

• [PyT] Add per-sequence attention policies to packed THD attention (#3…

  completed successfully:
  Deploy nightly docs#2030:Commit2a435a2pushedbyKshitijLakhani
  main

  Sep 3, 20261m 46s
  More actions for workflow run 2030