zenml

v0.94.5 Breaking

This release includes breaking changes for platform teams planning a safe upgrade.

Published 5d Model Serving & MLOps

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agentops agents ai automl data-science machine-learning

+12 more

devops-tools genai llm llmops metadata-tracking mlops pipelines production-ready pytorch tensorflow workflow zenml

Affected surfaces

auth

ReleasePort's take

Light signal

editorial:auto 5d

ZenML 0.94.5 adds real‑time event streaming for pipeline runs and a new statistics endpoint, improves observability with structured logging and OpenTelemetry support, and fixes several bugs including user account activation and GCP Cloud Run deployment timeouts.

Why it matters: Provides immediate operational visibility (streaming + stats) and reliability improvements; critical bugfixes resolve indefinite run states and account management failures in deployments using ZenML server.

Summary

AI summary

Broad release touches 🎨 Dashboard Improvements, 🛠️ Enhanced Developer Experience, https://github.com/zenml-io/zenml/pull/4804, and SSE.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Medium	Adds POST `/api/v1/runs/statistics` endpoint for aggregated pipeline run metrics. Adds POST `/api/v1/runs/statistics` endpoint for aggregated pipeline run metrics. Source: llm_adapter@2026-05-29 Confidence: high	—
Feature	Medium	Enhances Weights & Biases integration to auto‑add ZenML metadata and record identifiers bidirectionally. Enhances Weights & Biases integration to auto‑add ZenML metadata and record identifiers bidirectionally. Source: llm_adapter@2026-05-29 Confidence: high	—
Feature	Medium	Adds agentic human‑in‑the‑loop pipeline example demonstrating dynamic planning, approval gates, and lineage tracking. Adds agentic human‑in‑the‑loop pipeline example demonstrating dynamic planning, approval gates, and lineage tracking. Source: llm_adapter@2026-05-29 Confidence: high	—
Feature	Medium	Introduces structured logging and OpenTelemetry instrumentation for ZenML server. Introduces structured logging and OpenTelemetry instrumentation for ZenML server. Source: llm_adapter@2026-05-29 Confidence: high	—
Feature	Medium	Adds support for standard OpenTelemetry environment variables to configure ZenML server OTel settings. Adds support for standard OpenTelemetry environment variables to configure ZenML server OTel settings. Source: llm_adapter@2026-05-29 Confidence: high	—
Feature	Medium	Adds real-time event streaming for pipeline runs via `zenml.streaming.publish()`. Adds real-time event streaming for pipeline runs via `zenml.streaming.publish()`. Source: llm_adapter@2026-05-29 Confidence: low	—
Feature	Low	Enables real-time event streaming from pipelines using `zenml.streaming.publish()`. Enables real-time event streaming from pipelines using `zenml.streaming.publish()`. Source: granite4.1:30b@2026-05-29-audit Confidence: low	—
Feature	Low	Allows mounting a build cache during Docker image building for faster Python package installations. Allows mounting a build cache during Docker image building for faster Python package installations. Source: granite4.1:30b@2026-05-29-audit Confidence: low	—
Feature	Low	Supports string type annotations in step and pipeline definitions using quotes or `from __future__ import annotations`. Supports string type annotations in step and pipeline definitions using quotes or `from __future__ import annotations`. Source: granite4.1:30b@2026-05-29-audit Confidence: low	—
Feature	Low	Improves interactive wait condition input to accept raw strings without requiring quotes. Improves interactive wait condition input to accept raw strings without requiring quotes. Source: granite4.1:30b@2026-05-29-audit Confidence: low	—
Feature	Low	Adds configuration option `ZENML_CONSOLE_LOGGING_FORMAT` to control console output format (console, json, text). Adds configuration option `ZENML_CONSOLE_LOGGING_FORMAT` to control console output format (console, json, text). Source: granite4.1:30b@2026-05-29-audit Confidence: low	—
Bugfix
Bugfix	Medium	Handles keyboard interrupts during active wait conditions more gracefully, avoiding unclear error messages when server stops. Handles keyboard interrupts during active wait conditions more gracefully, avoiding unclear error messages when server stops. Source: llm_adapter@2026-05-29 Confidence: high	—
Bugfix	Medium	Corrects copy‑paste bug in user update endpoint that prevented activation/deactivation of accounts. Corrects copy‑paste bug in user update endpoint that prevented activation/deactivation of accounts. Source: llm_adapter@2026-05-29 Confidence: high	—
Bugfix	Medium	Fixes GCP Cloud Run deployer timeout for internal/authenticated services. Fixes GCP Cloud Run deployer timeout for internal/authenticated services. Source: llm_adapter@2026-05-29 Confidence: high	—
Bugfix	Medium	Fails pipeline runs when user pipeline function cannot be imported, preventing indefinite `running` state. Fails pipeline runs when user pipeline function cannot be imported, preventing indefinite `running` state. Source: llm_adapter@2026-05-29 Confidence: low	—
Refactor	Low	Deprecates `ZENML_LOGGING_FORMAT` in favor of `ZENML_CONSOLE_LOGGING_FORMAT` for console log formatting. Deprecates `ZENML_LOGGING_FORMAT` in favor of `ZENML_CONSOLE_LOGGING_FORMAT` for console log formatting. Source: granite4.1:30b@2026-05-29-audit Confidence: low	—

Full changelog

🚀 Live Event Streaming for Pipeline Runs

You can now stream custom events in real-time from your running pipelines! Call zenml.streaming.publish() from inside any step or dynamic pipeline to push events that can be consumed via Server-Sent Events (SSE). Enable this feature by setting stream_broker_implementation_source in your server configuration. The initial implementation includes a Redis-based broker with automatic catch-up, gap signaling, and idle cleanup. PR #4804

📊 Pipeline Run Statistics Endpoint

A new POST /api/v1/runs/statistics endpoint lets you query aggregated metrics across your pipeline runs. Group by status, pipeline, stack, user, time buckets (hour/day/week/month), metadata values, tags, and more. Calculate averages, sums, min/max over duration, step counts, cached steps, output artifacts, or custom numeric metadata. Perfect for building dashboards and analytics. PR #4860

🎯 Richer Weights & Biases Integration

The W&B experiment tracker now automatically adds ZenML pipeline and step metadata to your W&B runs, groups runs by pipeline execution, and records W&B identifiers back to ZenML step metadata. You can now configure custom groups, job types, run configs, explicit or deterministic run IDs, resume behavior, and pass through additional wandb.init kwargs for complete control over your experiment tracking. PR #4838

🤖 Agentic Human-in-the-Loop Pipeline Example

A new example demonstrates building dynamic agentic pipelines with human approval gates. The example shows how to plan agent tasks, fan them out with step.map(), summarize results, pause execution with zenml.wait() for human review, and branch the final action based on the decision. Includes clean lineage tracking with Annotated step outputs. PR #4849

📝 Structured Logging and OpenTelemetry Instrumentation

The ZenML server now supports structured logging with OpenTelemetry instrumentation. Configure console output with the new ZENML_CONSOLE_LOGGING_FORMAT environment variable, choosing between console, json, or text formats. Server logs use a clean structured layout with timestamps, levels, logger context, and optional JSON fields. PR #4781

🛠️ Enhanced Developer Experience

String Type Annotations: Step and pipeline definitions now support string annotations, either explicitly quoted or via from __future__ import annotations. PR #4843
Build Cache Mounting: Python package installations during Docker image building can now mount a build cache for faster builds. PR #4820
Improved Wait Condition Input: When resolving wait conditions interactively with string schemas, you can now input raw strings without quotes—ZenML detects and handles this automatically. PR #4845
Better Logging Control: Restored custom console log formatting for non-DEBUG output, added step-name prefixes in terminal output during execution (disable with ZENML_DISABLE_STEP_NAMES_IN_LOGS=true), and kept stored logs clean and unformatted. The ZENML_LOGGING_FORMAT variable is now deprecated in favor of ZENML_CONSOLE_LOGGING_FORMAT. PR #4851

🎨 Dashboard Improvements

Improved scrolling behavior for the pipeline timeline view. PR #1053

Fixed

Pipeline runs now properly fail when the user pipeline function cannot be imported (e.g., due to missing packages), instead of staying in a running state indefinitely. PR #4832
Keyboard interrupts during active wait conditions are now handled more gracefully, preventing unclear error messages when the server transitions to STOPPED status. PR #4835
Fixed a copy-paste bug in the user update endpoint where admins couldn't activate or deactivate user accounts—the code was incorrectly writing is_admin instead of active to the safe update object. PR #4839

What's Changed

Add version 0.94.3 to legacy docs by @github-actions[bot] in https://github.com/zenml-io/zenml/pull/4828
Document the organization trusted domain feature by @stefannica in https://github.com/zenml-io/zenml/pull/4819
Add 0.94.4 to the migration tests by @github-actions[bot] in https://github.com/zenml-io/zenml/pull/4829
Add option to mount build cache during python package installation by @schustmi in https://github.com/zenml-io/zenml/pull/4820
Fix CLI page info formatting by @strickvl in https://github.com/zenml-io/zenml/pull/4834
Fail a pipeline run if failing to import the user pipeline function by @schustmi in https://github.com/zenml-io/zenml/pull/4832
Only run DB backup/restore tests on local code version by @schustmi in https://github.com/zenml-io/zenml/pull/4831
Improved keyboard interrupt handling in active wait conditions by @schustmi in https://github.com/zenml-io/zenml/pull/4835
Bump the minor-and-patch group across 1 directory with 5 updates by @dependabot[bot] in https://github.com/zenml-io/zenml/pull/4847
Support string type annotations by @schustmi in https://github.com/zenml-io/zenml/pull/4843
Add agentic human-in-the-loop pipeline example by @htahir1 in https://github.com/zenml-io/zenml/pull/4849
Better string parsing during interactive wait condition resolving by @schustmi in https://github.com/zenml-io/zenml/pull/4845
Fix CLI tests in CI by @schustmi in https://github.com/zenml-io/zenml/pull/4850
Fixes for the logging env variables by @bcdurak in https://github.com/zenml-io/zenml/pull/4851
Fix GCP Cloud Run deployer timeout for internal/authenticated services by @safoinme in https://github.com/zenml-io/zenml/pull/4700
Structured logs and OTel Instrumentation for ZenML Server by @amitvikramraj in https://github.com/zenml-io/zenml/pull/4781
made shutdown_otel function idempotent and added support for individually enabling otel signals by @amitvikramraj in https://github.com/zenml-io/zenml/pull/4857
Bump actions/github-script from 8.0.0 to 9.0.0 by @dependabot[bot] in https://github.com/zenml-io/zenml/pull/4750
Allow streaming events while running a pipeline by @schustmi in https://github.com/zenml-io/zenml/pull/4804
fix: scoped sqlalchemy otel instrumentation to server sql engine by @amitvikramraj in https://github.com/zenml-io/zenml/pull/4858
fix(users): correct copy-paste bug in active status update by @sebastiondev in https://github.com/zenml-io/zenml/pull/4839
feat: supporting standard OTel env vars for zenml-server OTel configuration by @amitvikramraj in https://github.com/zenml-io/zenml/pull/4861
Richer WanDB integration by @Json-Andriopoulos in https://github.com/zenml-io/zenml/pull/4838
Run statistics endpoint by @schustmi in https://github.com/zenml-io/zenml/pull/4860
Prepare release 0.94.5 by @github-actions[bot] in https://github.com/zenml-io/zenml/pull/4871

New Contributors

@sebastiondev made their first contribution in https://github.com/zenml-io/zenml/pull/4839

Full Changelog: https://github.com/zenml-io/zenml/compare/0.94.4...0.94.5

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track zenml

Get notified when new releases ship.

About zenml

ZenML : One AI Platform from Pipelines to Agents. https://zenml.io.

All releases →