trainer
https://github.com/kubeflow/trainer
Go
Distributed AI Model Training and LLM Fine-Tuning on Kubernetes
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 trainer
Help out
- Issues
- chore(test): assert info state mutations and build errors in TestFlux- Fixes #3409
- [Bug/Feature Request] MPI MLPolicy does not propagate TrainJob.Spec.Trainer.Env to worker processes
- TrainJob STATE printer column shows Suspended after resume
- AI Scaffolding: Add .agents skill for add-training-runtime
- feat: promote TrainJobStatus feature gate to Beta
- chore(test): add JobSet plugin UTs for Status, IdentifyPodNetwork, and ReconcilerBuilders
- feat: KEP-2839 Dynamic LLM Trainer Framework
- feat: add MLX distributed E2E test for MPI workload
- Volcano PodGroup priorityClassName is set from the last ReplicatedJob, silently discarding the others
- Support GRPO-style post-training in Kubeflow Trainer via TrainingRuntime and targeted Torch plugin support
- Docs
- Go not yet supported