All workflows

Workflow runs

• [Metal] Add affine qmv_mma for 3-16 rows

  requires action with the application:
  Build and Test#5321:Pull request#4787synchronizebyahuzmeza
  ahuzmeza:pr/affine-qmv-mma-main

  Nov 2, 2026-1s
  More actions for workflow run 5321

• [CUDA] fix the cache key in cuda graph

  currently running:
  Build and Test#5320:Pull request#4821openedbynastya236
  cuda-graph-cache-key

  Nov 2, 2026In progress
  More actions for workflow run 5320

• [CUDA] top left mask for sdpa backward

  completed successfully:
  Build and Test#5319:Pull request#4820synchronizebynastya236
  attention-backward-mask

  Nov 2, 20261h 26m 0s
  More actions for workflow run 5319

• [CUDA] top left mask for sdpa backward

  failed:
  Build and Test#5318:Pull request#4820openedbynastya236
  attention-backward-mask

  Nov 2, 202628s
  More actions for workflow run 5318

• Validate trace input ids match trace inputs in import_function

  requires action with the application:
  Build and Test#5317:Pull request#4790synchronizebyturingH
  turingH:fix-mlxfn-input-count-mismatch

  Nov 2, 2026Action required
  More actions for workflow run 5317

• github_actions in /. - Update #1673493860

  completed successfully:
  Dependabot Updates#52:bydependabot[bot]Bot
  main

  Nov 2, 202646s

• Release build

  completed successfully:
  Release build#335:Scheduled
  main

  Nov 2, 202623m 56s
  More actions for workflow run 335

• Update bypass list

  completed successfully:
  Update bypass list#475:Scheduled
  main

  Nov 2, 202618s
  More actions for workflow run 475

• Pass the caller stream through complex l2_norm and scatter max/min VJP

  requires action with the application:
  Build and Test#5316:Pull request#4818openedbyAugustas11
  Augustas11:stream-complex-norm-scatter-vjp

  Nov 2, 2026Action required
  More actions for workflow run 5316

• [CUDA] Fix gather_mm with sorted_indices=True

  requires action with the application:
  Build and Test#5315:Pull request#4771synchronizebyluciusacarr
  luciusacarr:fix-gather-mm-sorted-lda

  Nov 2, 2026Action required
  More actions for workflow run 5315

• [CUDA] Fix gather_mm with sorted_indices=True

  requires action with the application:
  Build and Test#5314:Pull request#4771reopenedbyluciusacarr
  luciusacarr:fix-gather-mm-sorted-lda

  Nov 2, 2026Action required
  More actions for workflow run 5314

• [docs] Correct the global_scale_x argument name in qqmm

  requires action with the application:
  Build and Test#5313:Pull request#4817openedbyRedempt1onzzZZ
  Redempt1onzzZZ:codex/docs-qqmm-scale-4815

  Nov 2, 2026Action required
  More actions for workflow run 5313

• [docs] Mark keyword-only parameters in operation signatures

  requires action with the application:
  Build and Test#5312:Pull request#4816openedbyRedempt1onzzZZ
  Redempt1onzzZZ:codex/docs-keyword-only-4814

  Nov 2, 2026Action required
  More actions for workflow run 5312

• Fix missing API documentation and incorrect docstrings

  requires action with the application:
  Build and Test#5311:Pull request#4813openedbyXXXXRT666
  XXXXRT666:docs/api-coverage-audit

  Nov 1, 2026Action required
  More actions for workflow run 5311

• Add custom metal kernel for bilinear upsampling

  requires action with the application:
  Build and Test#5310:Pull request#4812openedbystarkAhmed43
  starkAhmed43:upsample-fused

  Nov 1, 2026Action required
  More actions for workflow run 5310

• Keep double precision in CPU exp, sin, cos, erf (#4789)

  completed successfully:
  Update bypass list#474:Commita8f66b5pushedbyzcbenz
  main

  Nov 1, 202618s
  More actions for workflow run 474

• Keep double precision in CPU exp, sin, cos, erf (#4789)

  completed successfully:
  Build and Test#5309:Commita8f66b5pushedbyzcbenz
  main

  Nov 1, 20262h 0m 51s
  More actions for workflow run 5309

• Check output size for 64-bit indexing in GPU SliceUpdate (#4794)

  completed successfully:
  Build and Test#5308:Commit9d5b21dpushedbyzcbenz
  main

  Nov 1, 20262h 7m 58s
  More actions for workflow run 5308

• Check output size for 64-bit indexing in GPU SliceUpdate (#4794)

  completed successfully:
  Update bypass list#473:Commit9d5b21dpushedbyzcbenz
  main

  Nov 1, 202616s
  More actions for workflow run 473

• Keep double precision in CPU exp, sin, cos, erf

  completed successfully:
  Build and Test#5307:Pull request#4789synchronizebyzcbenz
  mycomputercrashsalot:fix/cpu-float64-transcendentals

  Nov 1, 202630m 56s
  More actions for workflow run 5307

• [CPU] Fix accumulation precision in affine quantized matmul

  requires action with the application:
  Build and Test#5306:Pull request#4803synchronizebyerwinzhang7
  erwinzhang7:fix-cpu-qmm-bf16-accum

  Nov 1, 2026Action required
  More actions for workflow run 5306

• Add freqs and partial_rotary_factor support to RoPE (#4732)

  completed successfully:
  Update bypass list#472:Commit3d2fbcdpushedbyzcbenz
  main

  Nov 1, 202616s
  More actions for workflow run 472

• Add freqs and partial_rotary_factor support to RoPE (#4732)

  failed:
  Build and Test#5305:Commit3d2fbcdpushedbyzcbenz
  main

  Nov 1, 202652m 59s
  More actions for workflow run 5305

• [CUDA] Sequential kernel for gated_delta_update

  completed successfully:
  Build and Test#5304:Pull request#4783synchronizebyzcbenz
  cuda-gdn

  Nov 1, 20261h 44m 55s
  More actions for workflow run 5304

• [Metal] Add staged fp_qmm_t_nax with splitk and affine_qmv_nax kernel…

  completed successfully:
  Build and Test#5303:Commit9147324pushedbyzcbenz
  main

  Nov 1, 20261h 12m 33s
  More actions for workflow run 5303