Issues

• usage metering: streams cancelled after the last token aren't billed or alerted on

  needs: discussion 💬type: enhancement 💅🏼
  nebari-dev/llm-serving-pack#220 · Filter by author TimHunt12 opened on Oct 28, 2026

• usage metering: usage is lost while the collector or OpenMeter is down

  needs: discussion 💬type: enhancement 💅🏼
  nebari-dev/llm-serving-pack#219 · Filter by author TimHunt12 opened on Oct 28, 2026

• docs: usage metering gaps (namespace label, image tags, team attribution)

  area: documentation 📖
  nebari-dev/llm-serving-pack#218 · Filter by author TimHunt12 opened on Oct 28, 2026

• usage console: answer 503, not 401, when Keycloak's keys can't be refetched

  type: bug 🐛
  nebari-dev/llm-serving-pack#217 · Filter by author TimHunt12 opened on Oct 28, 2026

• shared/oidc: skip JWKS keys that aren't for signing (use: enc)

  type: bug 🐛
  nebari-dev/llm-serving-pack#216 · Filter by author TimHunt12 opened on Oct 28, 2026

• Limit the key manager's RBAC to the API-key Secrets and ConfigMaps it actually uses

  nebari-dev/llm-serving-pack#215 · Filter by author dcmcand opened on Oct 28, 2026

• Fix local dev setup Gateway API CRD channel conflict with Helm 4

  nebari-dev/llm-serving-pack#213 · Filter by author Adam-D-Lewis opened on Oct 22, 2026

• Add informational CA ConfigMap resolution status to PassthroughModel

  nebari-dev/llm-serving-pack#212 · Filter by author Adam-D-Lewis opened on Oct 22, 2026

• PassthroughModel routes have a hard-coded 120s request timeout that cannot be changed

  type: enhancement 💅🏼
  nebari-dev/llm-serving-pack#209 · Filter by author Adam-D-Lewis opened on Oct 16, 2026

• Plan routed coding-model providers and gateway readiness for the internal pilot

  nebari-dev/llm-serving-pack#200 · Filter by author trentoliphant opened on Sep 30, 2026

• Bound and validate PassthroughModel spec.provider.credentialSecretName

  priority: low 🌱type: enhancement 💅🏼
  nebari-dev/llm-serving-pack#199 · Filter by author dcmcand opened on Sep 28, 2026

• Add user-configurable service-level concurrency control for self-hosted models

  nebari-dev/llm-serving-pack#198 · Filter by author ns-mkusper opened on Sep 27, 2026
  

• Add opt-in native CPU KV-cache offload with explicit memory budgeting

  nebari-dev/llm-serving-pack#196 · Filter by author ns-mkusper opened on Sep 27, 2026

• Enforce per-caller token limits across served and external models

  nebari-dev/llm-serving-pack#195 · Filter by author ns-mkusper opened on Sep 24, 2026
  1

• Delete a model's routes and policies when the model is deleted

  nebari-dev/llm-serving-pack#194 · Filter by author ns-mkusper opened on Sep 23, 2026

• Advertise per-model context window on /v1/models

  needs: discussion 💬priority: low 🌱type: enhancement 💅🏼
  nebari-dev/llm-serving-pack#193 · Filter by author ns-mkusper opened on Sep 22, 2026

• Report each model's hosting origin on /v1/models

  nebari-dev/llm-serving-pack#191 · Filter by author ns-mkusper opened on Sep 22, 2026

• Add nodeSelector, tolerations and imagePullSecrets for the EPP under spec.advanced.inferencePool

  needs: discussion 💬
  nebari-dev/llm-serving-pack#176 · Filter by author tylerpotts opened on Aug 24, 2026

• Move operator to aigateway.envoyproxy.io/v1beta1 and update examples to Envoy Gateway v1.8.3 / AI Gateway v1.1.0

  type: deprecation ⚰️
  nebari-dev/llm-serving-pack#175 · Filter by author tylerpotts opened on Aug 24, 2026
  2

• Default EPP image ghcr.io/llm-d/llm-d-inference-scheduler:v0.8.0 no longer exists

  type: bug 🐛
  nebari-dev/llm-serving-pack#174 · Filter by author tylerpotts opened on Aug 24, 2026

• Add Playwright for E2E and Accessibility Testing

  priority: medium ⚡UI/UX/DX

  MFilter by Size: M

  MediumFilter by Priority: Medium
  Task

  nebari-dev/llm-serving-pack#152 · Filter by author jbouder opened on Aug 5, 2026
  1

• Emit privacy-safe OpenTelemetry traces from AI Gateway and vLLM

  nebari-dev/llm-serving-pack#148 · Filter by author ns-mkusper opened on Jul 27, 2026

• Internal JWT provider does not bind an audience by default

  priority: high 🔥securitytype: enhancement 💅🏼
  nebari-dev/llm-serving-pack#136 · Filter by author dcmcand opened on Jul 14, 2026

• Pin model-downloader init container to an immutable digest

  priority: medium ⚡securitytype: enhancement 💅🏼
  nebari-dev/llm-serving-pack#135 · Filter by author dcmcand opened on Jul 14, 2026

• LLMModel create/edit grants arbitrary container execution with Secret access

  priority: high 🔥securitytype: enhancement 💅🏼
  nebari-dev/llm-serving-pack#134 · Filter by author dcmcand opened on Jul 14, 2026