MLX

Workflow runs for MLX

• Allow overriding the QNN SDK version via buck config (#24102)

  completed successfully:
  MLX#5171:Commit95c40aepushedbymeta-codesync[bot]Bot
  main

  Oct 25, 202628s
  More actions for workflow run 5171

• Arm backend: Support constant negative pad values (#24108)

  completed successfully:
  MLX#5170:Commitd07c72bpushedbyErik-Lundell
  main

  Oct 24, 202645m 26s
  More actions for workflow run 5170

• MLX

  completed successfully:
  MLX#5169:bypytorch-bot[bot]Bot
  ciflow/mlx/24105

  Oct 24, 202651m 23s
  More actions for workflow run 5169

• MLX

  completed successfully:
  MLX#5168:bypytorch-bot[bot]Bot
  ciflow/mlx/24088

  Oct 24, 202645m 21s
  More actions for workflow run 5168

• Fetch Muse Glimmer serving tokenizer files from the non-GGUF repo (#2…

  completed successfully:
  MLX#5167:Commit185668bpushedbyGasoonjia
  main

  Oct 24, 202634s
  More actions for workflow run 5167

• Add group_independent_custom_shaders.py to the vgf BUCK target (#24138)

  completed successfully:
  MLX#5166:Commitea5cd99pushedbyGasoonjia
  main

  Oct 24, 202636s
  More actions for workflow run 5166

• MLX

  completed successfully:
  MLX#5165:bypytorch-bot[bot]Bot
  ciflow/mlx/24105

  Oct 24, 202646m 42s
  More actions for workflow run 5165

• MLX

  completed successfully:
  MLX#5164:bypytorch-bot[bot]Bot
  ciflow/mlx/24088

  Oct 24, 202645m 26s
  More actions for workflow run 5164

• Generate ring masks directly as packed bits (#24089)

  completed successfully:
  MLX#5163:Commit00f0654pushedbymeta-codesync[bot]Bot
  main

  Oct 24, 202636s
  More actions for workflow run 5163

• Support int16 in vision quantized_layer_norm (#24094)

  completed successfully:
  MLX#5162:Commit3bfbe82pushedbymeta-codesync[bot]Bot
  main

  Oct 24, 202647m 25s
  More actions for workflow run 5162

• CUDA wheel: check every dependency, and give torch_cpu_library target…

  completed successfully:
  MLX#5161:Commit7d9eb71pushedbyshoumikhin
  main

  Oct 23, 202640s
  More actions for workflow run 5161

• Make the @claude PR review a panel of three independent reviewers (#2…

  completed successfully:
  MLX#5160:Commit8916832pushedbyshoumikhin
  main

  Oct 23, 202635s
  More actions for workflow run 5160

• Keep the caller's numpy when setting up the Core ML backend (#24090)

  completed successfully:
  MLX#5159:Commit867a600pushedbyshoumikhin
  main

  Oct 23, 202639s
  More actions for workflow run 5159

• Arm backend: Fix Silero VAD end-to-end example (#24117)

  completed successfully:
  MLX#5158:Commit63485d1pushedbyusamahz
  main

  Oct 23, 202655m 57s
  More actions for workflow run 5158

• Unroll the M loop in fp_accumulate_with_int8_weight

  completed successfully:
  MLX#5157:Commit8d962f6pushedbymeta-codesync[bot]Bot
  main

  Oct 23, 202631s
  More actions for workflow run 5157

• CUDA wheel: check every dependency, and give torch_cpu_library targets torch's compile settings back

  completed successfully:
  MLX#5156:Pull request#24087synchronizebyshoumikhin
  shoumikhin:cuda-wheel-dep-followups

  Oct 23, 202656m 0s
  More actions for workflow run 5156

• Support int16 in generic quantized_layer_norm (#24093)

  completed successfully:
  MLX#5155:Commit703a411pushedbymeta-codesync[bot]Bot
  main

  Oct 23, 202647s
  More actions for workflow run 5155

• CUDA wheel: check every dependency, and give torch_cpu_library targets torch's compile settings back

  completed successfully:
  MLX#5154:Pull request#24087synchronizebyshoumikhin
  shoumikhin:cuda-wheel-dep-followups

  Oct 23, 20261h 7m 39s
  More actions for workflow run 5154

• CUDA wheel: check every dependency, and give torch_cpu_library targets torch's compile settings back

  completed successfully:
  MLX#5153:Pull request#24087synchronizebyshoumikhin
  shoumikhin:cuda-wheel-dep-followups

  Oct 23, 20261h 16m 53s
  More actions for workflow run 5153

• Support int16 in HiFi quantized_layer_norm (#24091)

  completed successfully:
  MLX#5152:Commitd533adfpushedbymeta-codesync[bot]Bot
  main

  Oct 23, 202641s
  More actions for workflow run 5152

• Arm backend: Restore Vela bounds checks with PAL initialization (#24115)

  completed successfully:
  MLX#5151:Commit82b7d19pushedbyusamahz
  main

  Oct 23, 20261h 9m 16s
  More actions for workflow run 5151

• Avoid copying the whole file when reading .pte and .ptd headers

  completed successfully:
  MLX#5150:Pull request#23561synchronizebyHiroForYou
  HiroForYou:avoid-header-copy

  Oct 23, 20261h 23m 38s
  More actions for workflow run 5150

• CUDA wheel: check every dependency, and give torch_cpu_library targets torch's compile settings back

  completed successfully:
  MLX#5149:Pull request#24087synchronizebyshoumikhin
  shoumikhin:cuda-wheel-dep-followups

  Oct 23, 202658m 6s
  More actions for workflow run 5149

• Arm backend: Support dynamic negative pad values (#24084)

  completed successfully:
  MLX#5148:Commit0295293pushedbyErik-Lundell
  main

  Oct 23, 202636s
  More actions for workflow run 5148

• CUDA wheel: check every dependency, and give torch_cpu_library targets torch's compile settings back

  completed successfully:
  MLX#5147:Pull request#24087synchronizebyshoumikhin
  shoumikhin:cuda-wheel-dep-followups

  Oct 23, 202647m 41s
  More actions for workflow run 5147