Ross ROSS = Recommend OSS · open-source software intelligence for agents

kubeflow/kubeflow

Machine Learning Toolkit for Kubernetes observed · 2026-08-28

github.com/kubeflow/kubeflow · homepage · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

84/100

  • Activity 98
  • Release rhythm 57
  • Longevity 100
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 152
  • age_days: 3198
  • days_rel: 126
  • days_push: 12
  • n_releases_24m: 4

Full methodology

Adoption not part of the score

15832 stars · 2690 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Kubeflow is a CNCF-graduated, Kubernetes-native toolkit for building AI/ML platforms, comprising subprojects like Pipelines, Notebooks, Trainer, Katib, Hub, and a Spark Operator. This repository is the gateway to those subprojects and shared metadata, with development happening in individual subproject repos.

Use cases

  • run machine learning pipelines on kubernetes
  • fine-tune llms on a kubernetes cluster
  • run jupyter notebooks in kubernetes
  • hyperparameter tuning and automl on k8s
  • orchestrate distributed model training with pytorch
  • build an end-to-end mlops platform
  • deploy rag and genai workflows at scale
  • manage model versions and registry metadata

When to choose

  • you already run Kubernetes and want a modular, composable AI platform on top of it
  • you need scalable distributed training, pipelines, notebooks, and hyperparameter search in one ecosystem
  • you want a CNCF-graduated, vendor-neutral foundation to build a custom AI platform
  • you need GenAI lifecycle tooling from synthetic data to inference and evaluation

When to avoid

  • you have no Kubernetes cluster or want a simple single-node ML setup
  • you need a lightweight all-in-one ML tool without significant operational overhead
  • you only need one small capability and don't want to deploy a full distribution
  • your team lacks Kubernetes administration experience

Facets

framework · maturity stable

machine-learning llm-training workflow-automation scheduling container-orchestration developer-tools machine-learning deep-learning large-language-models data-science cloud-computing cloud self-hosted python go mlops kubeflow pipelines notebooks hyperparameter-tuning distributed-training automl cncf genai model-registry containers devops kubernetes docker

10 sources

Member repositories

RepositoryRoleHealth v2
kubeflow/kubeflowmain84

For agents

markdown · JSON · MCP: product_card(name="kubeflow/kubeflow")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem