pre-commit

Workflow runs for pre-commit

• Clean up gfx1250 smoke test and bench serve params

  failed:
  pre-commit#2736:Pull request#1407synchronizebyafriedri
  afriedri:vllm-smoketest

  Oct 29, 20263m 59s
  More actions for workflow run 2736

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2735:Pull request#1415synchronizebyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 202610s
  More actions for workflow run 2735

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2734:Pull request#1415synchronizebyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 20262s
  More actions for workflow run 2734

• Enable Meta muse spark mxfp4

  skipped:
  pre-commit#2733:Pull request#1416openedbyJadenMathias
  jm/muse-gfx1250

  Oct 29, 20261s
  More actions for workflow run 2733

• [ROCm][Bugfix] W4A16 dense cache: count cached allocator blocks as free in the budget check

  completed successfully:
  pre-commit#2732:Pull request#1411synchronizebyzohaibm-amd
  zohaibm.w4a16-dequant-budget-empty-cache

  Oct 29, 20268m 38s
  More actions for workflow run 2732

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2731:Pull request#1415synchronizebyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 20261s
  More actions for workflow run 2731

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2730:Pull request#1415synchronizebyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 20262s
  More actions for workflow run 2730

• [ROCm][Bugfix] W4A16 dense cache: count cached allocator blocks as free in the budget check

  completed successfully:
  pre-commit#2729:Pull request#1411synchronizebyzohaibm-amd
  zohaibm.w4a16-dequant-budget-empty-cache

  Oct 29, 20266m 48s
  More actions for workflow run 2729

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2728:Pull request#1415synchronizebyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 20261s
  More actions for workflow run 2728

• [Model] Gemma 4: native packed vision tower on MMEncoderAttention

  skipped:
  pre-commit#2727:Pull request#1415openedbyroberteg16
  rogarcia.gemma4-native-vision

  Oct 29, 20261s
  More actions for workflow run 2727

• Merge upstream→gfx11

  completed successfully:
  pre-commit#2726:Pull request#1413openedbyeble-amd
  eble.merge-from-upstream

  Oct 28, 20266m 23s
  More actions for workflow run 2726

• [CI] Exclude nvidia-cuda-cccl from the ROCm test lockfile

  completed successfully:
  pre-commit#2725:Pull request#1412openedbyzohaibm-amd
  zohaibm.rocm-lockfile-exclude-cccl

  Oct 28, 202610m 9s
  More actions for workflow run 2725

• [ROCm][gfx1151] ROCM_RDNA_ATTN for multimodal encoders (non-causal)

  failed:
  pre-commit#2724:Pull request#1400synchronizebyroberteg16
  rogarcia.rdna-attn-non-causal-mm-encoder

  Oct 28, 20269m 21s
  More actions for workflow run 2724

• [ROCm][gfx1151] ROCM_RDNA_ATTN for multimodal encoders (non-causal)

  cancelled:
  pre-commit#2723:Pull request#1400synchronizebyroberteg16
  rogarcia.rdna-attn-non-causal-mm-encoder

  Oct 28, 20263s
  More actions for workflow run 2723

• [ROCm][Bugfix] W4A16 dense cache: count cached allocator blocks as free in the budget check

  failed:
  pre-commit#2722:Pull request#1411openedbyzohaibm-amd
  zohaibm.w4a16-dequant-budget-empty-cache

  Oct 28, 20266m 12s
  More actions for workflow run 2722

• Enable Kimi-K3 for Gfx1250

  skipped:
  pre-commit#2721:Pull request#1405synchronizebyJadenMathias
  jm/kimi

  Oct 28, 20261s
  More actions for workflow run 2721

• Enable Kimi-K3 for Gfx1250

  skipped:
  pre-commit#2720:Pull request#1405synchronizebyJadenMathias
  jm/kimi

  Oct 28, 20262s
  More actions for workflow run 2720

• Enable Kimi-K3 for Gfx1250

  skipped:
  pre-commit#2719:Pull request#1405synchronizebyJadenMathias
  jm/kimi

  Oct 28, 20261s
  More actions for workflow run 2719

• [ROCm][Perf] W4A16: opt-in dense weight cache for RDNA hybrid prefill

  completed successfully:
  pre-commit#2718:Pull request#1410synchronizebyzohaibm-amd
  port/prefill-dequant-cache

  Oct 28, 20269m 47s
  More actions for workflow run 2718

• [ROCm][Perf] W4A16: opt-in dense weight cache for RDNA hybrid prefill

  completed successfully:
  pre-commit#2717:Pull request#1410openedbyzohaibm-amd
  port/prefill-dequant-cache

  Oct 27, 202610m 7s
  More actions for workflow run 2717

• Merge upstream→gfx11

  completed successfully:
  pre-commit#2716:Pull request#1409openedbyeble-amd
  eble.merge-from-upstream

  Oct 27, 20266m 25s
  More actions for workflow run 2716

• [ROCm][RDNA][Perf] Fix MiniCPM-V TTFT regression from always-on ViT attention mask

  completed successfully:
  pre-commit#2715:Pull request#1398synchronizebystoivonen-amd
  stephant/idefics2-skip-encoder-mask-when-unpadded

  Oct 27, 20267m 0s
  More actions for workflow run 2715

• [ROCm][BugFix] Fix Qwen3-30B-A3B FP16 decode/TTFT regression from lost MoE weight padding

  completed successfully:
  pre-commit#2714:Pull request#1408openedbystoivonen-amd
  stephant/rocm-restore-moe-padding

  Oct 27, 20267m 54s
  More actions for workflow run 2714

• Enable Kimi-K3 for Gfx1250

  skipped:
  pre-commit#2713:Pull request#1405synchronizebyJadenMathias
  jm/kimi

  Oct 27, 202614s
  More actions for workflow run 2713

• [ROCm][gfx1151] ROCM_RDNA_ATTN for multimodal encoders (non-causal)

  completed successfully:
  pre-commit#2712:Pull request#1400synchronizebyroberteg16
  rogarcia.rdna-attn-non-causal-mm-encoder

  Oct 27, 20268m 8s
  More actions for workflow run 2712