publish-docker

Workflow runs for publish-docker

• perf(pytorch): limit KPool JIT specialization with bounded scoring (#…

  waiting:
  publish-docker#1865:Commit3f99ed1pushedbyRunningLeon
  main

  Oct 27, 2026Waiting
  More actions for workflow run 1865

• fix(serve): allow full-vocabulary chat logprobs with PyTorch (#5064)

  waiting:
  publish-docker#1864:Commit7ef6731pushedbysilencelamb
  main

  Oct 26, 2026Waiting
  More actions for workflow run 1864

• fix(gpt-oss): Reject failed GPT-OSS grammars for mandatory tool choic…

  waiting:
  publish-docker#1863:Commit7998cdbpushedbysilencelamb
  main

  Oct 24, 2026Waiting
  More actions for workflow run 1863

• feat(kvcache): support only save last checkpoint to reduce TTFT and p…

  waiting:
  publish-docker#1862:Commit54cd2d8pushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1862

• fix(serve): preserve GLM control-token literals (#5062)

  waiting:
  publish-docker#1861:Commit0810876pushedbysilencelamb
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1861

• fix(serve): correct output parsing, parameter forwarding, and EOS han…

  waiting:
  publish-docker#1860:Commite9262f8pushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1860

• fix(pytorch): unify prefix-cache lookahead for all MTP methods (#5061)

  waiting:
  publish-docker#1859:Commit2a920e6pushedbycaikun-pjlab
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1859

• feat(docker): update lmdeploy router download script (#5060)

  waiting:
  publish-docker#1858:Commitfef102epushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1858

• fix(serve): preserve request controls and response metadata (#5052)

  waiting:
  publish-docker#1857:Commitf91f103pushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1857

• perf(pytorch): optimize GLM-5.3 Flash (#5027)

  waiting:
  publish-docker#1856:Commit577ec3fpushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1856

• fix(serve): accept null image detail in Responses API (#5054)

  waiting:
  publish-docker#1855:Commit82d76ccpushedbyRunningLeon
  main

  Oct 23, 2026Waiting
  More actions for workflow run 1855

• fix(serve): preserve tool-result image ownership (#5044)

  waiting:
  publish-docker#1854:Commitaba987apushedbyRunningLeon
  main

  Oct 22, 2026Waiting
  More actions for workflow run 1854

• fix(serve): reject unsupported chat content blocks (#5047)

  waiting:
  publish-docker#1853:Commit2301681pushedbyRunningLeon
  main

  Oct 22, 2026Waiting
  More actions for workflow run 1853

• [Feature] Add optional FP8 MLA cache for GLM-5.3-Flash (#5045)

  waiting:
  publish-docker#1852:Commit1890f50pushedbyRunningLeon
  main

  Oct 21, 2026Waiting
  More actions for workflow run 1852

• Reset local prefix cache on sleep (#5035)

  waiting:
  publish-docker#1851:Commit2b3464epushedbyRunningLeon
  main

  Oct 21, 2026Waiting
  More actions for workflow run 1851

• Support original z-lab Qwen3 DFlash checkpoints; require checkpoint t…

  waiting:
  publish-docker#1850:Commit187870cpushedbyRunningLeon
  main

  Oct 21, 2026Waiting
  More actions for workflow run 1850

• fix(anthropic): retain tool parsing with raw input tokens (#5032)

  waiting:
  publish-docker#1849:Commite463b3dpushedbyRunningLeon
  main

  Oct 21, 2026Waiting
  More actions for workflow run 1849

• [Fix] Support user image inputs in Responses API (#5043)

  waiting:
  publish-docker#1848:Commit52e57d8pushedbysilencelamb
  main

  Oct 20, 2026Waiting
  More actions for workflow run 1848

• fix(deps): add lmdeploy-router and mooncake into docker (#5040)

  waiting:
  publish-docker#1847:Commit55d3101pushedbyRunningLeon
  main

  Oct 20, 2026Waiting
  More actions for workflow run 1847

• fix(pytorch): exclude GLM-5.3 runtime tails from prefix checkpoints (…

  waiting:
  publish-docker#1846:Commit56a6c79pushedbyRunningLeon
  main

  Oct 19, 2026Waiting
  More actions for workflow run 1846

• fix(pytorch): support GLM-5.3 Flash BF16 EP and R3 (#5036)

  waiting:
  publish-docker#1845:Commitcb3e021pushedbyRunningLeon
  main

  Oct 19, 2026Waiting
  More actions for workflow run 1845

• fix(disagg): await Mooncake async migration executor future (#4971) (…

  waiting:
  publish-docker#1844:Commitfa4b736pushedbyRunningLeon
  main

  Oct 19, 2026Waiting
  More actions for workflow run 1844

• [Fix] Finalize weight readiness after distributed updates (#5033)

  waiting:
  publish-docker#1843:Commit0a10113pushedbyRunningLeon
  main

  Oct 16, 2026Waiting
  More actions for workflow run 1843

• feat(kv_connector): mooncake store support GLM-5.3-Flash (#5034)

  waiting:
  publish-docker#1842:Commit30fd106pushedbyRunningLeon
  main

  Oct 16, 2026Waiting
  More actions for workflow run 1842

• [Fix] Honor FP32 LM head and update MTP weights for GLM-5.2 (#5003)

  waiting:
  publish-docker#1841:Commitd6764fcpushedbyRunningLeon
  main

  Oct 16, 2026Waiting
  More actions for workflow run 1841