37 versions

  • feat: inject PET_* envs into init containers via envInjection config (#3516)no API changese3c21902026-06-16 12:52
  • feat(api): Add terminationGracePeriodSeconds to PodSpecPatch in TrainJob (#3324)no API changes7c161d02026-05-26 17:36
  • feat: storageuri targetjob validation fix (#3510)no API changesfa2e7cf2026-05-12 21:25
  • feat: update k8s to 0.36 (#3470)1ce6a94f2026-05-08 12:38
  • feat: add support for tracking TrainJob progress and training metrics (#3227)no API changes66d0b0b2026-03-13 20:01
  • feat: add activeDeadlineSeconds (#3258)no API changes466c97a2026-03-13 18:37
  • feat(api): Replace PodTemplateOverrides with RuntimePatches API (#3309)438d82432026-03-12 16:29
  • feat(runtimes): Add XGBoost runtime(KEP-2598) (#3200)no API changesdb104792026-03-10 10:04
  • feat: support for flux framework as hpc manager (#3188)no API changes08b7a192026-03-10 03:19
  • feat(api): BREAKING CHANGE: Remove numProcPerNode from Torch API (#3239)no API changese6b83762026-03-01 06:53
  • feat(api): BREAKING CHANGE – Remove ElasticPolicy API (#3235)1abf42d52026-02-24 20:58
  • feat: Add the manager field to the podTemplateOverride object (#3020)no API changes99deac72026-02-16 23:54
  • feat(api): Fix immutability of the TrainJob APIs (#3157)no API changes4dc33172026-02-09 19:23
  • feat(runtimes): Add JAX training runtime (#3151)no API changes0fea3ec2026-02-06 18:15
  • chore(deps): Bump JobSet v0.11.0 and LWS v0.8.0 (#3144)no API changes09af1a12026-01-27 23:53
  • chore(deps): Bump Go 1.25, k8s v1.35, and controller-runtime v0.23.1 (#3127)15568ea22026-01-27 19:50
  • feat(api): Add securityContext support to PodTemplateSpecOverride in TrainJob (#3066)no API changesf9214832026-01-15 15:30
  • chore(ci): Enable Kubernetes API Linter (#2858)no API changes2732af62025-10-17 14:40
  • feat(api): Add PodTemplateOverrides API into TrainJob (#2785)2d56715c2025-10-09 10:08
  • fix: Allow multiple podSpec overrides to target the same TargetJob (#2880)no API changes25022bd2025-10-07 12:55
  • feat: KEP-2437 - PodGroup Creation for Volcano Scheduler (#2729)no API changes4e17ce52025-10-06 05:51
  • chore(operator): Upgrade JobSet to v0.10.1 (#2875)no API changesb1b597d2025-10-03 10:08
  • chore(operator): Bump JobSet to v0.9.0 version (#2821)no API changes0d3dd9e2025-09-09 06:55
  • feat: support imagePullSecrets in TrainJob pod spec overrides (#2806)no API changesf2e7e0d2025-09-08 12:43
  • chore(operator): Upgrade Kubernetes to v1.34 (#2804)no API changesa728b3f2025-08-28 17:02
  • feat: support affinity in TrainJob pod spec overrides (#2796)no API changesa350fa02025-08-28 15:52
  • Upgrade Kubernetes to v1.33 (#2756)no API changes45ce64f2025-07-28 15:18
  • feat: Add schedulingGates to PodSpecOverrides (#2700)no API changes1ea27122025-07-03 11:03
  • feat(controller): Implement PodSpecOverride API (#2614)no API changes68d4ba12025-06-06 17:04
  • feat(controller): Refactor the Initializer APIs of TrainJob (#2523)4b323b442025-03-17 07:15
  • feat(sdk): Migrate to OpenAPI V3 (#2490)no API changes7e0b2ab2025-03-10 13:45
  • Implement MPI numProcPerNode defaulter (#2483)no API changes0cb910a2025-03-06 22:48
  • feat(sdk): Generate external Kubernetes and JobSet models (#2466)no API changes35a1c3b2025-03-05 19:08
  • KEP-2170: Add validation to Torch `numProcPerNode` field (#2409)no API changesea42d112025-02-14 20:57
  • KEP-2170: Implement MPI Plugin for Kubeflow Trainer (#2394)no API changes3f3a8d32025-02-14 17:55
  • Update the naming conventions for Kubeflow Trainer (#2415)39430603322025-02-06 13:48
  • unversionedlatest2026-07-17 12:04