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
- bug: manageRoleReplicasPerGroup panics on unowned pod with empty OwnerReferences, crashing kthena-controller-manager
- Router JWT auth stays broken for 7 days after a brief JWKS outage at startup
- bug(router): TokenRateLimiter consumes input tokens before output check and leaves stale side limiters on partial update
- spec.plugins[].scope.target is unvalidated and has no default, the kubebuilder markers are not applied
- Improper error wrapping: fmt.Errorf uses `%v` instead of `%w` across the codebase
- bug(router): HTTPConnector and NIXLConnector store per-request state on connector struct, causing data races and request corruption under concurrency
- ModelBooster controller never starts on a workload only install
- fix(model-serving): recover worker goroutines from pod processing panics
- fix(router): clear stale rate limiter when ModelRoute rateLimit is removed
- Role readiness check uses the latest configuration for partition-protected ServingGroups
- Docs
- Go not yet supported