All pull requests

Pull requests

• feat(sidecar): collect sidecar runtime and native engine metrics for vLLM, SGLang and TRT-LLM

  #620·nv-kmcgill53 opened Oct 22, 2026Collaborator·

  ·
  1

• fix(power): publish module power through a transient GPU disagreement

  #619·edwingao28 opened Oct 22, 2026Contributor·

  ·

• fix: refuse a loopback address when srun IP resolution fails

  #611·edwingao28 opened Oct 16, 2026Contributor·

  ·
  1

• fix(atom): size -tp from the worker's GPUs over its PP and PCP stages

  #604·Jasen2201 opened Oct 14, 2026·

  ·
  2

• feat(power): read x86 RAPL package power with --source rapl

  #603·edwingao28 opened Oct 14, 2026Contributor·

  ·
  1

• feat(benchmark): add HLE eval script for NeMo Skills

  #601·AlphaBladez opened Oct 12, 2026Collaborator·

  ·

• fix: allow configurable infrastructure startup waits

  #593·anish-shanbhag opened Oct 9, 2026·

  ·

• fix(reporting): inherit cluster S3 defaults with recipe overrides

  #581·nealvaidya opened Oct 6, 2026·

  ·
  2

• fix(vllm): stop disabling plugins in sidecar mode

  #578·alec-flowers opened Oct 6, 2026Collaborator·

  ·

• feat: Launcher ABC; launcher: docker runs a job on a single GPU machine without Slurm

  #567·ishandhanani opened Oct 4, 2026Collaborator·

  ·
  3

  

• feat(power): standalone HTML visualizer for scrape-timings.jsonl

  #558·FrankD412 opened Oct 1, 2026Collaborator·

  ·
  1

• feat(telemetry): enable DCGM power collection by default

  devin
  #550·peilii opened Sep 30, 2026Collaborator·

  ·

• feat(reporting): support installed result publisher commands

  #542·nv-yna opened Sep 29, 2026Collaborator·

  ·
  1

• feat(dsight): attach configuration lineage to runtime metrics

  #521·nv-yna opened Sep 28, 2026Collaborator·

  ·
  1

• enable torch profiler

  #505·shengf-nv opened Sep 24, 2026·

  ·

• feat(dsight): join Dynamo request and worker identities

  #502·nv-yna opened Sep 23, 2026Collaborator·

  ·
  1

• fix(slurm): share one Pyxis container per image per node across srun steps

  devin
  #484·Tabrizian opened Sep 21, 2026Member·

  ·
  12

• fix(power): isolate endpoint sampling overruns

  #480·Ankur-singh opened Sep 18, 2026Collaborator·

  ·
  5

• feat(power-report): interactive Pareto / power-over-time HTML report (AIP-1388)

  #475·FrankD412 opened Sep 17, 2026Collaborator·

  ·
  1

• refactor(logs): fan the run log directory out into workers/, services/logs/, telemetry/, fingerprints/

  #469·FrankD412 opened Sep 16, 2026Collaborator·

  ·
  1

• fix: capture native engine and sidecar metrics with Tachometer

  #465·alec-flowers opened Sep 16, 2026Collaborator·

  ·

• fix: collect Dynamo sidecar metrics for custom benchmarks

  #464·alec-flowers opened Sep 16, 2026Collaborator·

  ·
  1

• feat(RL): OpenEnv terminal-bench rollouts without Docker

  #461·ishandhanani opened Sep 16, 2026Collaborator·

  ·

• feat(dashboard): generate raw Tachometer UI for TRT-LLM runs

  #447·nv-yna opened Sep 14, 2026Collaborator·

  ·
  1

• fix(frontend): use Dynamo's two-tier worker-selection policy for KV routing

  #444·ishandhanani opened Sep 14, 2026Collaborator·

  ·
  1