-
Notifications
You must be signed in to change notification settings - Fork 40
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
[Feature]: Support queue per tenant merge policy
enhancementNew feature or requestNew feature or requestStatus: Open.[Feature]: Add GCP PubSub Support to the Producer Library
enhancementNew feature or requestNew feature or requestStatus: Open.#438 In llm-d/llm-d-async;[Feature]: Cancel inference requests that are already executing
enhancementNew feature or requestNew feature or requestStatus: Open.#422 In llm-d/llm-d-async;- Status: Open.#421 In llm-d/llm-d-async;
[Feature]: Feedback-based dispatch control (aimd pool gate driven by response feedback)
enhancementNew feature or requestNew feature or requestStatus: Open.- Status: Open.
transports: per-queue/per-message result queue routing for redis-pubsub and gcp-pubsub (parity with redis-sortedset)
enhancementNew feature or requestNew feature or requestStatus: Open.#416 In llm-d/llm-d-async;tier-priority: nothing enforces that reserved-lane objectives outrank overflow-lane objectives (inversion silently defeats the quota model)
bugSomething isn't workingSomething isn't workingdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.The router recipe's base values do not fit an 8-vCPU node
documentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#357 In llm-d/llm-d-async;[Feature]: Consider active requests and inflight tokens for saturation signal
enhancementNew feature or requestNew feature or requestStatus: Open.[Feature]: Throttle Worker Pool Workers by Gate Budget
enhancementNew feature or requestNew feature or requestStatus: Open.[Feature]: Response body-transform extension point (symmetric to #255) for post-dispatch payload rewriting
enhancementNew feature or requestNew feature or requestStatus: Open.