All pull requests

Pull requests

• fix: keep Hugging Face Ray inference off the event loop

  deployLLMtests
  #791·0z5a opened Nov 4, 2026·

  ·
  1

  1

• fix: preserve generation settings across Ray batches

  deployLLMtests
  #789·0z5a opened Nov 4, 2026·

  ·
  1

  1

• Fix synchronous batch response latency reporting

  deployLLMscriptstests
  #787·0z5a opened Nov 4, 2026·

  ·
  1

• ci: update release docs during code freeze

  CI
  #784·balasaajay opened Oct 27, 2026Contributor·

  ·
  26

• fix(deploy): stop non-primary Ray ranks entering generate_other_ranks in coordinator mode

  bugdeployLLMqa_rcca_donetests
  #782·pruprakash opened Oct 26, 2026Contributor·

  ·
  4

  

• fix(deploy): pass trust_remote_code to vLLM exporter in HFRayDeployable

  deployLLMtests
  #781·v0ropaev opened Oct 20, 2026·

  ·
  1

• fix(deploy): handle echo without logprobs in Megatron Ray completions

  deployLLMtests
  #780·v0ropaev opened Oct 20, 2026·

  ·
  1

• build(security): require cryptography 50

  #777·balasaajay opened Oct 19, 2026Contributor·

  ·
  2

• Fix agent skills: normalize unofficial frontmatter fields per the Agent Skills spec

  exporttests
  #772·janEbert opened Oct 10, 2026·

  ·

• build: update wandb to 0.30.0

  #770·balasaajay opened Oct 8, 2026Contributor·

  ·
  1

• fix: combine CI matrix strategy and action description fixes

  CItests
  #760·andrewwhitecdw opened Aug 31, 2026Contributor·

  ·
  1

• ci: AUT-1606 bump release library to v1.8.10

  CI
  #753·svcnemo-autobot opened Aug 19, 2026Contributor·

  ·
  2

• Fix the bug with MG (#748)

  deployLLM
  #752·oyilmaz-nvidia opened Aug 19, 2026Contributor·

  ·
  2

• cp: fix(deploy): resolve inference coordinator host to bindable address (743) into r0.7.0

  cherry-pickdeployLLMRun CICDtests
  #751·svcnvidia-nemo-ci opened Aug 19, 2026Contributor·

  ·
  3

• chore: update claude review to FW-CI-templates v1.8.5

  CI
  #727·kajalj22 opened Jul 29, 2026Contributor·

  ·
  1

• fix: mark None-default parameters as Optional in vLLMExporter

  exporttestsvLLM
  #724·andrewwhitecdw opened Jul 24, 2026Contributor·

  ·
  6

• fix: allow None dynamic_axes in OnnxLLMExporter.export

  exportONNXtests
  #723·andrewwhitecdw opened Jul 24, 2026Contributor·

  ·
  3

• fix: use integer dtype for random_seed in query script

  deployLLMscriptstests
  #722·andrewwhitecdw opened Jul 24, 2026Contributor·

  ·
  3

• fix: use min_output_len value in query_llm inputs

  deployLLMtests
  #721·andrewwhitecdw opened Jul 24, 2026Contributor·

  ·
  3

• Test removal of decord and opencv-python-headless

  CIdeploytests
  #681·chtruong814 opened May 15, 2026Contributor·

  ·
  4

• Update MM docs

  deploydocumentationexportmultimodalr0.7.0TensorRT-LLM
  #498·meatybobby opened Oct 31, 2025Contributor·

  ·
  3