kthena
https://github.com/volcano-sh/kthena
Go
Lightweight, Modular, Kubernetes-native AI serving platform for scalable model serving.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Go not yet supported1 Subscribers
Add a CodeTriage badge to kthena
Help out
- Issues
- [Feature Request] Coordinate Prefill/Decode rolling-update progress with maxSkew
- Feat: add ModelServing topology affinity proposal
- runtime agent restart permanently drops KV cache ownership for blocks the engine still holds
- HTTPRoute can use a cross-namespace InferencePool without ReferenceGrant
- Compile HTTPRoute regex path matchers once instead of on every request
- fix(router): TokenRateLimiter input consume and partial clears
- runtime: retry Redis operations against the current client
- fix(autoscaler): reject minReplicas greater than maxReplicas for all target types
- Proposal: add adaptive rolling update granularity for ModelServing
- fix(downloader): preserve destination lock during PVC sync
- Docs
- Go not yet supported