Skip to content

Pull requests: ai-dynamo/dynamo

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
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

test(vllm): wait on a gate instead of a clock in the EP capacity timeout test backend::vllm Relates to the vllm backend external-contribution Pull request is from an external contributor size/M test
#15127 opened Sep 21, 2026 by glamr-agent Contributor Loading…
fix(sglang): fail fast when an mm-router worker dies during startup backend::sglang Relates to the sglang backend external-contribution Pull request is from an external contributor fix size/M
#15125 opened Sep 20, 2026 by glamr-agent Contributor Loading…
fix(runtime): stop FileStore creating buckets after shutdown external-contribution Pull request is from an external contributor fix size/L
#15124 opened Sep 20, 2026 by glamr-agent Contributor Loading…
test(sglang): cover thinking token budget size/M test
#15123 opened Sep 20, 2026 by biswapanda Contributor 3/3 Loading…
docs: explain thinking token budgets docs documentation Improvements or additions to documentation size/M
#15122 opened Sep 20, 2026 by biswapanda Contributor 2/3 Loading…
docs(sglang): add thinking budget request example backend::sglang Relates to the sglang backend docs size/S
#15121 opened Sep 20, 2026 by biswapanda Contributor 1/3 Loading…
feat(sglang): support thinking token budgets backend::sglang Relates to the sglang backend feat frontend `python -m dynamo.frontend` and `dynamo-run in=http|text|grpc` multimodal size/XL
#15120 opened Sep 20, 2026 by biswapanda Contributor Loading…
test(sglang): pin sampling on the parallel tool-call protocol test external-contribution Pull request is from an external contributor size/XS test
#15119 opened Sep 20, 2026 by glamr-agent Contributor Loading…
refactor(common-http): fetch media through one policy-applying entrypoint backend::sglang Relates to the sglang backend backend::trtllm Relates to the trtllm backend backend::vllm Relates to the vllm backend documentation Improvements or additions to documentation multimodal refactor size/L
#15118 opened Sep 20, 2026 by nnshah1 Contributor Loading…
test(trtllm): state the KV cache memory fraction on profiled launches backend::trtllm Relates to the trtllm backend external-contribution Pull request is from an external contributor size/L test
#15117 opened Sep 20, 2026 by glamr-agent Contributor Loading…
perf(sglang): enable push response egress backend::sglang Relates to the sglang backend backend::trtllm Relates to the trtllm backend external-contribution Pull request is from an external contributor perf size/M trusted-contributor Org-External user who is trusted to run CI without Org-member approval
#15116 opened Sep 20, 2026 by xianlubird Contributor Loading…
perf(frontend): avoid JSON round-trip for chunk metrics external-contribution Pull request is from an external contributor frontend `python -m dynamo.frontend` and `dynamo-run in=http|text|grpc` perf size/M trusted-contributor Org-External user who is trusted to run CI without Org-member approval
#15115 opened Sep 20, 2026 by xianlubird Contributor Loading…
1 task done
feat(sglang): accept routed_experts_start_len on chat/completions backend::sglang Relates to the sglang backend external-contribution Pull request is from an external contributor feat frontend `python -m dynamo.frontend` and `dynamo-run in=http|text|grpc` size/M
#15114 opened Sep 20, 2026 by DongBaiYue Loading…
docs: Dynamo v1.5.0 release notes page and release ledger docs documentation Improvements or additions to documentation size/XXL
#15113 opened Sep 20, 2026 by dagil-nvidia Collaborator Loading…
2 of 3 tasks
build(vllm): track nightly runtime image backend::vllm Relates to the vllm backend build container external-contribution Pull request is from an external contributor size/M
#15112 opened Sep 20, 2026 by sungsooha Contributor Draft
feat(vllm): record engine provenance in FPM self-benchmark artifacts backend::vllm Relates to the vllm backend documentation Improvements or additions to documentation feat size/XXL
#15110 opened Sep 20, 2026 by Arsene12358 Contributor Loading…
test(vllm): bound the gc_maintain freeze-count assertion backend::vllm Relates to the vllm backend external-contribution Pull request is from an external contributor size/XS test
#15109 opened Sep 19, 2026 by glamr-agent Contributor Loading…
docs: promote v1.5.0 to current release docs documentation Improvements or additions to documentation size/L
#15108 opened Sep 19, 2026 by dagil-nvidia Collaborator Loading…
3 of 4 tasks
test(vllm): allocate the self-benchmark disagg worker ports external-contribution Pull request is from an external contributor size/S test
#15107 opened Sep 19, 2026 by glamr-agent Contributor Loading…
fix(vllm): background the aggregated multimodal worker backend::vllm Relates to the vllm backend external-contribution Pull request is from an external contributor fix size/L
#15104 opened Sep 19, 2026 by glamr-agent Contributor Loading…
fix(responses): include cache write tokens in usage details fix frontend `python -m dynamo.frontend` and `dynamo-run in=http|text|grpc` size/L
#15103 opened Sep 19, 2026 by pskiran1 Contributor Draft
fix(llm): echo Responses API request metadata fix frontend `python -m dynamo.frontend` and `dynamo-run in=http|text|grpc` size/L
#15102 opened Sep 19, 2026 by pskiran1 Contributor Draft
fix(tests): pin SGLang router worker NCCL ports and gate on /health external-contribution Pull request is from an external contributor fix size/S
#15099 opened Sep 19, 2026 by glamr-agent Contributor Loading…
ProTip! What’s not been updated in a month: updated:<2026-08-20.