Issues

• docs: announce the 2026-10-26 move to dsx-ai-factory/nvcf in the README

  documentation
  NVIDIA/nvcf#2661 · Filter by author kristinapathak opened on Oct 16, 2026
  

• worker: investigate streamed request-body latency

  NVIDIA/nvcf#2659 · Filter by author FamousDirector opened on Oct 16, 2026
  

• vanity-gateway: report wrong JSON field types as invalid_type

  needs-triage
  NVIDIA/nvcf#2658 · Filter by author sachin-prabhakar opened on Oct 16, 2026

• gateways: validate caller-supplied X-Request-Id length and characters

  needs-triage
  NVIDIA/nvcf#2657 · Filter by author sachin-prabhakar opened on Oct 16, 2026
  1

• design: function status when the instance service is unreachable

  NVIDIA/nvcf#2650 · Filter by author shobham-nv opened on Oct 16, 2026

• NVCA operator: detect and repair webhook CA mismatches on existing clusters

  NVIDIA/nvcf#2649 · Filter by author shobham-nv opened on Oct 16, 2026
  

• Gate cluster visibility and allocation per NCA ID in ICMS

  needs-triage
  Enhancement

  NVIDIA/nvcf#2645 · Filter by author harshm98 opened on Oct 16, 2026
  

• grpc-proxy shutdown cuts off calls still waiting on a worker

  invocation-planeneeds-triage
  NVIDIA/nvcf#2643 · Filter by author shobham-nv opened on Oct 16, 2026

• Deduplicate overlapping producer release dispatches

  NVIDIA/nvcf#2641 · Filter by author sbaum1994 opened on Oct 16, 2026

• Clear replacement-scan and numeric-user gates for the umbrella patch

  NVIDIA/nvcf#2640 · Filter by author sbaum1994 opened on Oct 16, 2026

• Align initial producer release versions and release-tooling messages

  NVIDIA/nvcf#2633 · Filter by author sbaum1994 opened on Oct 16, 2026

• Repair automatic chart-bump reconciliation after concurrent component releases

  NVIDIA/nvcf#2632 · Filter by author sbaum1994 opened on Oct 16, 2026

• Qualify shared Dockerfile release runners before migrating producers

  NVIDIA/nvcf#2631 · Filter by author sbaum1994 opened on Oct 16, 2026

• Restore and verify OpenBao migration image publication

  NVIDIA/nvcf#2630 · Filter by author sbaum1994 opened on Oct 16, 2026

• Vanity Gateway: expose token limits and support authorized discovery of private models

  NVIDIA/nvcf#2629 · Filter by author bewestphal opened on Oct 16, 2026

• pylon: backends whose first token takes more than about 6 s never become routable

  NVIDIA/nvcf#2628 · Filter by author sachin-prabhakar opened on Oct 15, 2026

• test(bdd): delete functions and API keys at scenario teardown

  NVIDIA/nvcf#2627 · Filter by author sachin-prabhakar opened on Oct 15, 2026

• test(bdd): run every LLM scenario on the in-repo OpenAI-compatible sample

  NVIDIA/nvcf#2622 · Filter by author sachin-prabhakar opened on Oct 15, 2026

• Stabilize worker-utils service and worker test teardown on macOS

  NVIDIA/nvcf#2617 · Filter by author sbaum1994 opened on Oct 15, 2026
  1

• GitHub Actions skip CI on release/* stack branches

  NVIDIA/nvcf#2611 · Filter by author nvaghela-oss opened on Oct 15, 2026

• NVCA builds the BYOO sidecar from a stale vendored translation library

  bugcompute-stack
  Bug

  NVIDIA/nvcf#2604 · Filter by author kristinapathak opened on Oct 15, 2026

• Add BDD coverage for BYOO in self-managed deployments

  compute-stack
  Task

  NVIDIA/nvcf#2603 · Filter by author kristinapathak opened on Oct 15, 2026

• Document BYOO for self-managed deployments

  compute-stackdocumentation
  Task

  NVIDIA/nvcf#2602 · Filter by author kristinapathak opened on Oct 15, 2026

• Add a BYOO end-to-end sample function and reference telemetry sink

  compute-stack
  Task

  NVIDIA/nvcf#2601 · Filter by author kristinapathak opened on Oct 15, 2026

• Add telemetry endpoint management to nvcf-cli

  control-plane
  Task

  NVIDIA/nvcf#2600 · Filter by author kristinapathak opened on Oct 15, 2026