AI / ML ML Pipeline stable

Kubeflow

ML toolkit for Kubernetes-native model training and serving

14.0K stars 600 contributors Since 2018
Website → GitHub

End-to-end ML platform on Kubernetes with pipelines, hyperparameter tuning (Katib), model serving (KServe), Jupyter notebooks, and distributed training operators.

License
Apache-2.0
Min RAM
8 GB
Min CPUs
4 cores
Scaling
distributed
Complexity
expert
Performance
enterprise grade
Self-hostable
K8s native
Offline
Pricing
fully free
Docs quality
good
Vendor lock-in
none

Use cases

  • End-to-end ML pipelines on Kubernetes
  • Distributed training across GPU clusters
  • Hyperparameter search with Katib
  • Model serving with KServe at scale

Anti-patterns / when NOT to use

  • Requires Kubernetes — not for simple setups
  • Complex installation and maintenance
  • Overkill for single-model projects
  • Steep learning curve

Compare with alternatives

Replaces / alternatives to

  • SageMaker
  • Vertex AI
  • Azure ML

Technical specs

Language
PythonGo
API type
RESTSDK
Protocols
HTTP
Deployment
docker
SDKs
python

Community

GitHub stars 14.0K
Contributors 600
Commit frequency weekly
Plugin ecosystem none
Backing Google / CNCF
Funding foundation

Release

Latest version
Last release
Since 2018

Best fit

Team size
mediumenterprise
Industries
enterpriseresearchfintech

Tags

  • mlops
  • kubernetes
  • pipelines
  • hyperparameter-tuning
  • model-serving
  • distributed-training