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
- Surface generated workload failures in ModelBooster and ModelServing status
- bug: kthena-router does not remove rate limiters when spec.rateLimit is cleared from ModelRoute
- [enhancement] kvblocks accounting during the fairness queue
- Bug: Uncanceled Context goroutine leak due to context.TODO() in ModelServingController API calls
- router: kvcache GC scans one batch per hourly tick, a full pass takes days at modest keyspace sizes
- Disaggregated autoscaler can patch a ratio outside ratioConstraint after integer rounding
- ModelServing template metadata can overwrite controller-reserved Pod labels
- Kthena Release Tracking Project
- Scale-down interrupts in-flight requests: pods deleted without draining (propose draining annotation + onFlight-based two-phase delete)
- Autoscaler does not limit pod metric response size
- Docs
- Go not yet supported