All workflows

Workflow runs

• [CUDA] Sequential kernel for gated_delta_update

  failed:
  Build and Test#5216:Pull request#4783synchronizebyzcbenz
  cuda-gdn

  Oct 29, 202632m 57s
  More actions for workflow run 5216

• Build and test for macOS 27

  completed successfully:
  Build and Test#5215:Pull request#4786synchronizebyzcbenz
  macos27

  Oct 29, 202619m 44s
  More actions for workflow run 5215

• metal: read FP16 and BF16 affine constants directly for FP32 inputs

  requires action with the application:
  Build and Test#5214:Pull request#4785synchronizebybri-prism
  PrismML-Eng:upstream/mixed-affine-constants-20261028

  Oct 29, 2026Action required
  More actions for workflow run 5214

• Build and test for macOS 27

  completed successfully:
  Build and Test#5213:Pull request#4786synchronizebyzcbenz
  macos27

  Oct 28, 202630m 15s
  More actions for workflow run 5213

• [CUDA] Sequential kernel for gated_delta_update

  failed:
  Build and Test#5212:Pull request#4783synchronizebyzcbenz
  cuda-gdn

  Oct 28, 20261h 27m 16s
  More actions for workflow run 5212

• Fix gguf tensor to f16 size overflow (#4764)

  completed successfully:
  Update bypass list#449:Commit93007e9pushedbyzcbenz
  main

  Oct 28, 202617s
  More actions for workflow run 449

• Fix gguf tensor to f16 size overflow (#4764)

  completed successfully:
  Build and Test#5211:Commit93007e9pushedbyzcbenz
  main

  Oct 28, 202634m 25s
  More actions for workflow run 5211

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

  requires action with the application:
  Build and Test#5210:Pull request#4787openedbyahuzmeza
  ahuzmeza:pr/affine-qmv-mma-main

  Oct 28, 2026Action required
  More actions for workflow run 5210

• Release build

  failed:
  Release build#328:Manually runbyzcbenz
  macos27

  Oct 28, 20264m 25s
  More actions for workflow run 328

• Build and test for macOS 27

  completed successfully:
  Build and Test#5209:Pull request#4786openedbyzcbenz
  macos27

  Oct 28, 202623m 13s
  More actions for workflow run 5209

• [CUDA] Sequential kernel for gated_delta_update

  cancelled:
  Build and Test#5208:Pull request#4783synchronizebyzcbenz
  cuda-gdn

  Oct 28, 202636m 2s
  More actions for workflow run 5208

• metal: read FP16 and BF16 affine constants directly for FP32 inputs

  requires action with the application:
  Build and Test#5207:Pull request#4785openedbybri-prism
  PrismML-Eng:upstream/mixed-affine-constants-20261028

  Oct 28, 2026Action required
  More actions for workflow run 5207

• Fix CUDA matmul and addmm reduction precision

  failed:
  Build and Test#5206:Pull request#4784openedbytetsuo
  tetsuo:fix-cuda-reduction-precision

  Oct 28, 202614h 39m 56s
  More actions for workflow run 5206

• [CUDA] Sequential kernel for gated_delta_update

  completed successfully:
  Build and Test#5205:Pull request#4783openedbyzcbenz
  cuda-gdn

  Oct 28, 20261h 37m 29s
  More actions for workflow run 5205

• Fix FP8 NaN preservation in to_fp8 and E8M0 scale dequantization (#4756)

  completed successfully:
  Build and Test#5204:Commit7d2d94apushedbyzcbenz
  main

  Oct 28, 20261h 38m 6s
  More actions for workflow run 5204

• Fix FP8 NaN preservation in to_fp8 and E8M0 scale dequantization (#4756)

  completed successfully:
  Update bypass list#448:Commit7d2d94apushedbyzcbenz
  main

  Oct 28, 202621s
  More actions for workflow run 448

• [Metal] Add staged fp_qmm_t_nax with splitk and affine_qmv_nax kernels

  requires action with the application:
  Build and Test#5203:Pull request#4781openedbyXXXXRT666
  XXXXRT666:metal-nax-quantized-mid-rows

  Oct 28, 2026Action required
  More actions for workflow run 5203

• Fix Metal expm1 argument reduction accuracy

  requires action with the application:
  Build and Test#5202:Pull request#4780openedbyarcusbuilds
  arcusbuilds:fix/expm1-ln2-correction

  Oct 28, 2026Action required
  More actions for workflow run 5202

• Fix Metal and CUDA affine quantization for all-negative groups (#4775)

  completed successfully:
  Update bypass list#447:Commit32a183epushedbyzcbenz
  main

  Oct 28, 202619s
  More actions for workflow run 447

• Fix Metal and CUDA affine quantization for all-negative groups (#4775)

  completed successfully:
  Build and Test#5201:Commit32a183epushedbyzcbenz
  main

  Oct 28, 20262h 54m 11s
  More actions for workflow run 5201

• Fix FP8 NaN preservation in to_fp8 and E8M0 scale dequantization

  completed successfully:
  Build and Test#5200:Pull request#4756synchronizebyzcbenz
  FaizPalwala:fix-fp8-nan-handling

  Oct 28, 20261h 24m 28s
  More actions for workflow run 5200

• Fix GPU concatenate for outputs larger than 2^31 elements (#4758)

  completed successfully:
  Build and Test#5199:Commitdf46f58pushedbyzcbenz
  main

  Oct 28, 20261h 26m 46s
  More actions for workflow run 5199

• Fix GPU concatenate for outputs larger than 2^31 elements (#4758)

  completed successfully:
  Update bypass list#446:Commitdf46f58pushedbyzcbenz
  main

  Oct 28, 202617s
  More actions for workflow run 446

• Release build

  completed successfully:
  Release build#327:Scheduled
  main

  Oct 28, 202617m 25s
  More actions for workflow run 327

• Update bypass list

  completed successfully:
  Update bypass list#445:Scheduled
  main

  Oct 28, 202616s
  More actions for workflow run 445