{"full_name":"zenml-io/zenml","owner":"zenml-io","name":"zenml","shard":"1","url":"https://github.com/zenml-io/zenml","description":"ZenML 🙏: One AI Platform from Pipelines to Agents. https://zenml.io.","homepage":"https://zenml.io","topics":["mlops","machine-learning","data-science","production-ready","devops-tools","zenml","pipelines","metadata-tracking","deep-learning","pytorch","tensorflow","ml","ai","automl","workflow"],"tag_pages":["ai","automl","data-science","deep-learning","devops-tools","machine-learning","ml","mlops","pytorch","tensorflow","workflow"],"language":"Python","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/88676955?v=4","created_at":"2020-11-19T09:25:46Z","pushed_at":"2026-11-13T15:28:38Z","metrics":{"stars":5636,"forks":674,"watchers":42,"issues":715,"prs":4696,"releases":169,"issues_open":134,"issues_closed":581,"prs_open":42,"prs_closed":603,"prs_merged":4051,"discussions":14},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[5639,5639,5639,5639,5635,5636,5636],"forks":[674,674,674,674,674,674,674],"watchers":[null,42,42,42,42,42,42],"issues_open":[null,134,134,134,134,134,134],"prs_open":[null,41,39,39,41,42,42]},"boards":{},"same_owner":[],"issues":[{"number":2532,"title":"Integrate `safetensors` for model serialization","comments":23,"created_at":"2024-03-15T10:46:43Z"},{"number":2219,"title":"Create Documentation on Unit Testing ZenML Steps","comments":22,"created_at":"2024-01-03T13:43:59Z"},{"number":2217,"title":"Integrate Kubernetes Connector Support with Kaniko Image Builder","comments":14,"created_at":"2024-01-03T13:17:13Z"},{"number":2262,"title":"Resolve Vertex Scheduler Character Limit Issue for Long Pipeline Names","comments":13,"created_at":"2024-01-10T16:42:05Z"},{"number":2218,"title":"Enhance Vertex AI Orchestrator for Compatibility with Private IP Cloud SQL","comments":8,"created_at":"2024-01-03T13:23:22Z"},{"number":4468,"title":"Display failure/success hooks and their output","comments":6,"created_at":"2026-02-05T16:56:36Z"},{"number":4398,"title":"Adding support for Sagemaker v3","comments":6,"created_at":"2026-01-14T14:08:36Z"},{"number":4209,"title":"Create fallback logo for custom components","comments":6,"created_at":"2025-11-18T09:48:46Z"}],"discussions":[{"number":5384,"title":"Follow-up on Vulnerability Report Submitted 3 Weeks Ago","comments":1,"category":"Q & A","created_at":"2026-10-25T12:15:56Z"},{"number":5370,"title":"Monitoring tab for the dashboard: would an outside contribution be welcome?","comments":1,"category":"Forum","created_at":"2026-10-18T14:50:27Z"},{"number":5369,"title":"ZenML CLI included in AgentManager's tool catalog","comments":0,"category":"Show and tell","created_at":"2026-10-18T11:50:48Z"},{"number":5314,"title":"Decision-level regression testing between ML pipeline versions with DeciShift","comments":2,"category":"Show and tell","created_at":"2026-09-25T18:18:52Z"},{"number":4936,"title":"MCP to help OSS devs understand zenml-io/zenml","comments":0,"category":"Show and tell","created_at":"2026-06-12T12:02:48Z"},{"number":4794,"title":"Roadmap question: OpenTelemetry metrics export for ZenML server","comments":7,"category":"Forum","created_at":"2026-05-02T17:05:45Z"},{"number":4787,"title":"RFC: First-class LLM evaluation as a stack component (`evaluator` type)","comments":3,"category":"Forum","created_at":"2026-04-30T02:36:03Z"},{"number":235,"title":"[CLOSED] Which model explainability tool to integrate with Q1 2022","comments":1,"category":"Show and tell","created_at":"2021-12-14T13:48:30Z"}],"latest_release":{"name":"0.98.1","tag":"0.98.1","published_at":"2026-11-04T05:40:29Z","body":"<!-- ZENML_GITBOOK_RELEASE_NOTES_START tag=0.98.1 -->\n- **Rename projects from the CLI and SDK**: You can now rename projects with `zenml project update <name-or-uuid>` using `--name`, `--display-name`, and `--description` ([PR #5367](https://github.com/zenml-io/zenml/pull/5367)). Technical names and display names can be changed independently while preserving the project UUID and associated resources, and SDK updates now correctly preserve omitted fields or explicitly clear descriptive fields.\n- **Helm chart pods now satisfy restricted Pod Security defaults**: The default ZenML Helm chart pod security settings now work in namespaces enforcing Kubernetes Pod Security `restricted` ([PR #5397](https://github.com/zenml-io/zenml/pull/5397)). This also applies the chart security values consistently to CA certificate init containers and Helm test pods, reducing the extra overrides needed for locked-down clusters.\n\n<details>\n<summary>Fixed</summary>\n\n- **More reliable integration installs**: Installing ZenML integrations is now less likely to break ZenML’s already-installed dependency set ([PR #5399](https://github.com/zenml-io/zenml/pull/5399)). This prevents cases where adding an integration upgrades shared packages into incompatible versions and causes import failures before pipelines can run.\n- **GCP runner OpenTelemetry compatibility**: GCP runner images now avoid installing an incompatible OpenTelemetry API version alongside ZenML’s pinned SDK ([PR #5366](https://github.com/zenml-io/zenml/pull/5366)). This fixes import failures that could happen before any pipeline steps started.\n- **MLflow integration installs on Python 3.13 and slim images**: The MLflow integration no longer depends on the unused `python-rapidjson` package ([PR #5360](https://github.com/zenml-io/zenml/pull/5360)). This avoids source compilation failures in environments without build tools such as `g++`, especially on Python 3.13 or slim container images.\n- **GCS-backed runs no longer hang during final uploads or cleanup**: ZenML aligned GCS-related dependencies and added optional storage controls to avoid pipeline steps getting stuck after their work and outputs have already completed ([PR #5373](https://github.com/zenml-io/zenml/pull/5373)). This addresses hangs from indefinite GCS responses and affected `gcsfs` cleanup behavior, helping orchestrators observe completed containers instead of waiting forever.\n- **Step outputs are preserved after execution or finalization failures**: ZenML no longer deletes step output directories from the launcher exception handler when a step fails after artifacts or server-side completion have already been recorded ([PR #5362](https://github.com/zenml-io/zenml/pull/5362)). Persisted metadata will no longer point to missing files in these partial-failure scenarios, while the original error is still propagated.\n- **Replay runs include defaulted pipeline parameters**: Replaying a run now forwards pipeline parameters that were satisfied by defaults in the original run ([PR #5361](https://github.com/zenml-io/zenml/pull/5361)). This keeps replayed runs consistent with the source run even when users did not explicitly pass every parameter.\n- **CLI labels can contain equals signs**: CLI label parsing now splits `--label` values only on the first `=`, so values such as URLs and query strings are accepted ([PR #5383](https://github.com/zenml-io/zenml/pull/5383)). This fixes label handling across service connector and stack component commands where values like `https://example.com/?a=b` were previously rejected or misinterpreted.\n- **Hosted Classic dashboard CORS origins are allowed automatically**: Pro workspace servers now automatically include the hosted Classic dashboard origin in addition to their configured dashboard origin ([PR #5396](https://github.com/zenml-io/zenml/pull/5396)). Production and staging Classic dashboard URLs are matched exactly over HTTPS, while custom enterprise dashboards and existing custom CORS origins keep ","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}