Skip to content

Pull requests: llm-d/llm-d-router

New pull request New
110 Open 1,555 Closed
110 Open 1,555 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

docs: document that endpoint discovery honours custom pod readinessGates area/docs area/epp kind/documentation Categorizes issue or PR as related to documentation. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2604 opened Aug 28, 2026 by almogtavor Loading…
1
1
fix(chart): allow overriding the EPP Deployment strategy so multi-replica EPP can satisfy helm --wait kind/bug Categorizes issue or PR as related to a bug. size/M Denotes a PR that changes 30-99 lines, ignoring generated files.
#2603 opened Aug 28, 2026 by almogtavor Loading…
1
1
Fix: Scope Endpoint Metric Cleanup by Namespace and Producer area/epp area/scheduling kind/bug Categorizes issue or PR as related to a bug.
#2602 opened Aug 28, 2026 by weizhoublue Contributor Loading…
4
Fix KV event hash integer conversion area/kvcache size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2601 opened Aug 28, 2026 by Alex-ai-future Contributor Loading…
5 tasks done
4
Add audio content block support to tokenizer for multimodal routing area/epp area/kvcache area/scheduling size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2600 opened Aug 27, 2026 by RishabhSaini Contributor Loading…
1
docs: add release-note fragment for PR #2458 size/XS Denotes a PR that changes 0-9 lines, ignoring generated files.
#2597 opened Aug 27, 2026 by llm-d-router-release-notes Bot Loading…
feat(epp): add OTel span to prefix cache affinity filter decisions area/epp area/scheduling size/M Denotes a PR that changes 30-99 lines, ignoring generated files.
#2594 opened Aug 27, 2026 by sudoalok Contributor Loading…
1
fix(request): strip min_tokens in single-token request priming area/coordinator area/sidecar kind/bug Categorizes issue or PR as related to a bug. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2593 opened Aug 27, 2026 by roytman Contributor Loading…
4 tasks done
1
5
fix(sidecar): forward untouched request fields as raw JSON to keep key order area/sidecar kind/bug Categorizes issue or PR as related to a bug. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2591 opened Aug 27, 2026 by rebel-jinhwan Loading…
2 tasks done
1
6
refactor(sidecar): extract shared concurrent prefill/decode dispatch helper area/sidecar kind/cleanup size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2589 opened Aug 27, 2026 by roytman Contributor Loading…
2 tasks done
1
6
Credit shared KV tiers via node:/pool: pseudo-pod identities area/datalayer area/epp area/kvcache area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
#2586 opened Aug 27, 2026 by rebel-jinhwan • Draft
4 tasks done
2
feat(datalayer): confine per-request attribute access to declarations area/datalayer area/epp area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
#2580 opened Aug 26, 2026 by satyamg1620 Contributor • Draft
1
1
feat(epp): gate DisaggregatedSet readiness on initial Pod snapshot area/datalayer area/epp area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2576 opened Aug 26, 2026 by hasB4K Contributor Loading…
3
tracing: support OTLP/HTTP as an alternative to the gRPC exporter area/telemetry kind/feature Categorizes issue or PR as related to a new feature. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2573 opened Aug 26, 2026 by satyamg1620 Contributor Loading…
9 tasks done
1
1
feat(coordinator): share revision decisions across phases [Safe DisaggSet 4/4] area/coordinator area/datalayer area/docs area/epp area/scheduling size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2572 opened Aug 26, 2026 by hasB4K Contributor Loading…
1 task done
feat(epp): coordinate DisaggregatedSet revision decisions [Safe DisaggSet 2/4] area/coordinator area/datalayer area/docs area/epp area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2571 opened Aug 26, 2026 by hasB4K Contributor Loading…
1 task done
1
feat(epp): add atomic cross-replica coordination API [Safe DisaggSet 1/4] area/datalayer area/epp kind/feature Categorizes issue or PR as related to a new feature. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
#2570 opened Aug 26, 2026 by hasB4K Contributor Loading…
2
feat(flowcontrol): add program-score ordering policy area/epp area/flowcontrol size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
#2565 opened Aug 26, 2026 by elinacse Contributor Loading…
1
fix(epp): preserve endpoint attributes in data-parallel profile area/epp area/scheduling kind/bug Categorizes issue or PR as related to a bug. size/S Denotes a PR that changes 10-29 lines, ignoring generated files.
#2563 opened Aug 26, 2026 by yuanhuizang-cloud Contributor Loading…
1
3
Route cached follow-ups by remaining prefill work area/datalayer area/epp area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files.
#2559 opened Aug 26, 2026 by nilig Collaborator Loading…
5 tasks done
1
test(requestcontrol): give each Admit sub-test its own reqCtx kind/bug Categorizes issue or PR as related to a bug. size/XS Denotes a PR that changes 0-9 lines, ignoring generated files.
#2549 opened Aug 25, 2026 by elinacse Contributor Loading…
1
1
[ROCm] Add ATOM connector area/sidecar size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
#2537 opened Aug 25, 2026 by vcave Loading…
4
feat(epp): add session-state-producer plugin area/datalayer area/epp area/scheduling kind/feature Categorizes issue or PR as related to a new feature. size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
#2531 opened Aug 24, 2026 by paulzhn Loading…
1
4
feat(sidecar): support sglang rust native p/d area/docs area/sidecar size/S Denotes a PR that changes 10-29 lines, ignoring generated files.
#2525 opened Aug 24, 2026 by sagearc Contributor • Draft
perf(epp): optimize HTTP token ID request parsing area/epp kind/cleanup size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files.
#2522 opened Aug 24, 2026 by yankay Contributor Loading…
5 of 6 tasks
1
12
ProTip! What’s not been updated in a month: updated:<2026-07-28.