xTag

镜像概览

完整引用
docker.io/kubeflow/training-operator
Registry
docker.io
命名空间
kubeflow
仓库
training-operator
已缓存 Tag 数
422
最近同步
07/13/2026, 07:57 PM

简介

Kubeflow Training Operator 镜像，管理 PyTorchJob/TFJob/MPIJob 等分布式训练 CR，K8s ML 训练任务生命周期控制器。

详细介绍

kubeflow/training-operator 运行 Kubeflow Training Operator，监听 Training Job CR 创建分布式 PyTorch/TensorFlow/MPI 训练 Pod 组。定位为 K8s ML 训练 Operator 控制器，需 GPU 节点与 RBAC；大规模分布式模型训练、Kubeflow Pipelines 训练步骤或 MLOps 自动化训练调度。

快速启动

docker pull docker.io/kubeflow/training-operator:v1.8.0
# kubectl apply -f training-operator manifests

推荐实践

# 创建 PyTorchJob CR 指定 worker/replica
# Operator 自动管理 Pod 生命周期

核心参数说明

  • CR — PyTorchJob/TFJob/MPIJob
  • GPU — 训练 Pod 需 nvidia.com/gpu
  • RBAC — 集群级控制器权限
  • 勿 — 独立 docker run 训练任务

最新 Tag（20 条）

查看全部 422 个 Tag →
Tag架构推送时间大小Digest同步期望操作
docker.io/kubeflow/training-operator
:v1-8e41c03
linux/amd64linux/arm64linux/ppc64le
2026/04/30 23:3330.5 MBsha256:a6cc929e1b3d…
docker.io/kubeflow/training-operator
:latest
linux/amd64linux/arm64linux/ppc64le
2026/04/30 23:3330.5 MBsha256:a6cc929e1b3d…
docker.io/kubeflow/training-operator
:v1-9121e6a
linux/amd64linux/arm64linux/ppc64le
2026/04/10 22:1230.5 MBsha256:d64c4337fffc…
docker.io/kubeflow/training-operator
:v1-9e12c68
linux/amd64linux/arm64linux/ppc64le
2026/01/10 00:4030.5 MBsha256:0b409bddc8cb…
docker.io/kubeflow/training-operator
:v1-52e41d5
linux/amd64linux/arm64linux/ppc64le
2025/07/18 00:2330.4 MBsha256:417f82454849…
docker.io/kubeflow/training-operator
:v1-c77ee3f
linux/amd64linux/arm64linux/ppc64le
2025/07/17 22:3830.4 MBsha256:ab21867dce39…
docker.io/kubeflow/training-operator
:v1-d6eb98e
linux/amd64linux/arm64linux/ppc64le
2025/07/17 10:5430.4 MBsha256:a656fd59b0d4…
docker.io/kubeflow/training-operator
:v1-d104c19
linux/amd64linux/arm64linux/ppc64le
2025/07/17 02:5030.4 MBsha256:daa28122e961…
docker.io/kubeflow/training-operator
:v1-800bd77
linux/amd64linux/arm64linux/ppc64le
2025/06/30 20:0830.4 MBsha256:5ef56457f8cc…
docker.io/kubeflow/training-operator
:v1-8c29c7b
linux/amd64linux/arm64linux/ppc64le
2025/06/20 16:3730.4 MBsha256:d30a95cfc39a…
docker.io/kubeflow/training-operator
:v1-d022585
linux/amd64linux/arm64linux/ppc64le
2025/05/03 22:2630.4 MBsha256:62faecfb10a1…
docker.io/kubeflow/training-operator
:v1-bde9c20
linux/amd64linux/arm64linux/ppc64le
2025/05/03 11:0930.4 MBsha256:bd2ba866d62e…
docker.io/kubeflow/training-operator
:v1-3f15cb8
linux/amd64linux/arm64linux/ppc64le
2025/05/03 06:0030.4 MBsha256:1d5c0fa0abc5…
docker.io/kubeflow/training-operator
:v1-b9cb87e
linux/amd64linux/arm64linux/ppc64le
2025/05/02 21:1130.4 MBsha256:4857ea550b5b…
docker.io/kubeflow/training-operator
:v1-f58e893
linux/amd64linux/arm64linux/ppc64le
2025/04/30 00:5130.4 MBsha256:c1786b1fa5ec…
docker.io/kubeflow/training-operator
:v1-378a9e6
linux/amd64linux/arm64linux/ppc64le
2025/04/08 01:3130.4 MBsha256:0e5b47e7a567…
docker.io/kubeflow/training-operator
:v1-77bd5cd
linux/amd64linux/arm64linux/ppc64le
2025/03/29 09:0630.4 MBsha256:fa244972af00…
docker.io/kubeflow/training-operator
:v1-5c0e763
linux/amd64linux/arm64linux/ppc64le
2025/03/25 03:3430.4 MBsha256:fda431bfcc44…
docker.io/kubeflow/training-operator
:v1-31da2d6
linux/amd64linux/arm64linux/ppc64le
2025/03/25 01:4630.4 MBsha256:7f72fad31c30…
docker.io/kubeflow/training-operator
:v1-5840e81
linux/amd64linux/arm64linux/ppc64le
2025/03/19 16:5730.4 MBsha256:0d934ea91874…