Add Modal orchestrator with step operator and orchestrator flavors #3733

htahir1 · 2025-06-12T09:36:41Z

New Features

Introduced Modal orchestrator integration, enabling orchestration and scaling of ML pipelines on Modal's serverless cloud platform with support for both pipeline and per-step execution modes.
Added configuration options for resource allocation, authentication, environment selection, and execution modes for both orchestrator and step operator.

Documentation

Added comprehensive guides for the new Modal orchestrator and enhanced the Modal step operator documentation with setup, configuration, authentication, and usage best practices.
Updated component guide and table of contents to include Modal orchestrator details and clarified differences between orchestrator and step operator.

Refactor

Modularized step operator logic by moving validation, image management, and authentication into shared utility functions for improved maintainability.

Tests

Updated tests to align with the new parameter structure for GPU configuration utilities.

Breaking Changes

None - this is a new integration that doesn't affect existing functionality.

Dependencies

Adds modal>=0.64.49,<1 requirement to Modal integration
No new dependencies for core ZenML

Note: This orchestrator follows the same patterns as other ZenML orchestrators
(GCP Vertex, Kubernetes) and integrates seamlessly with the existing ZenML
stack architecture.

Note: I also updated the step operator logic to unify it

Pre-requisites

Please ensure you have done the following:

I have read the CONTRIBUTING.md document.
I have added tests to cover my changes.
I have based my new branch on develop and the open PR is targeting develop. If your branch wasn't based on develop read Contribution guide on rebasing branch to develop.
IMPORTANT: I made sure that my changes are reflected properly in the following resources:
- ZenML Docs
- Dashboard: Needs to be communicated to the frontend team.
- Templates: Might need adjustments (that are not reflected in the template tests) in case of non-breaking changes and deprecations.
- Projects: Depending on the version dependencies, different projects might get affected.

Types of changes

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to change)
Other (add details above)

Summary by CodeRabbit

New Features

Introduced Modal orchestrator integration, enabling orchestration and scaling of ML pipelines on Modal's serverless cloud platform with configurable execution modes and resource settings.
Added comprehensive documentation for the Modal orchestrator, including setup, configuration, usage, troubleshooting, and best practices.

Enhancements

Expanded Modal step operator documentation with detailed setup instructions, authentication options, GPU support, and environment management.
Improved configuration options for the Modal step operator, including environment, timeout, and secure API token handling.

Bug Fixes

Updated internal references and tests to align with new configuration parameter structures.

Refactor

Modularized Modal integration by introducing utility functions for authentication, resource management, image handling, and stack validation.

Documentation

Updated and expanded documentation for Modal orchestrator and step operator, including a new entry in the component guide table of contents.

coderabbitai · 2025-06-12T09:36:48Z

Important

Review skipped

Auto reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

📝 Walkthrough

Walkthrough

This update introduces a new Modal orchestrator integration to ZenML, enabling orchestration of entire pipelines or individual steps on Modal's serverless cloud. It adds new configuration, settings, and utility modules, updates the Modal integration to support both orchestrator and step operator flavors, enhances documentation, and refactors related code for modularity and clarity.

Changes

Files/Paths	Change Summary
docs/book/component-guide/orchestrators/README.md, docs/book/component-guide/orchestrators/modal.md, docs/book/component-guide/step-operators/modal.md, docs/book/component-guide/toc.md	Documentation updated and expanded for Modal orchestrator and step operator; new orchestrator docs and TOC entry added.
src/zenml/integrations/modal/init.py	Modal integration updated to support orchestrator flavor, new constant, requirements updated, and flavors method expanded.
src/zenml/integrations/modal/flavors/init.py	Public exports for Modal orchestrator config, flavor, and settings added.
src/zenml/integrations/modal/flavors/modal_orchestrator_flavor.py	New orchestrator flavor, config, settings, and execution mode enum introduced.
src/zenml/integrations/modal/flavors/modal_step_operator_flavor.py	Step operator config and settings expanded with new authentication, environment, and timeout options.
src/zenml/integrations/modal/orchestrators/init.py	New module exposing the ModalOrchestrator class.
src/zenml/integrations/modal/orchestrators/modal_orchestrator.py	ModalOrchestrator class implemented for running pipelines on Modal, with stack validation and execution logic.
src/zenml/integrations/modal/orchestrators/modal_orchestrator_entrypoint.py	Entrypoint for Modal orchestrator sandbox: parses args, dispatches execution, handles modes, finalizes runs.
src/zenml/integrations/modal/orchestrators/modal_orchestrator_entrypoint_configuration.py	Entrypoint configuration class for orchestrator CLI invocation and argument construction.
src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py	ModalSandboxExecutor class added to manage pipeline/step execution, resource config, image caching, and sandbox lifecycle.
src/zenml/integrations/modal/step_operators/modal_step_operator.py	Refactored to use shared utilities for stack validation, Modal client setup, image building, and GPU value extraction.
src/zenml/integrations/modal/utils.py	New utility module: Modal authentication, resource/image helpers, stack validator, and tagging functions added.
tests/integration/integrations/modal/step_operators/test_modal_step_operator.py	Test updated to match new signature of get_gpu_values utility function.

Sequence Diagram(s)

sequenceDiagram
    participant User
    participant ZenML
    participant ModalOrchestrator
    participant ModalSandboxExecutor
    participant Modal Cloud

    User->>ZenML: Trigger pipeline run
    ZenML->>ModalOrchestrator: Prepare deployment
    ModalOrchestrator->>ModalSandboxExecutor: Setup execution (pipeline/step mode)
    ModalSandboxExecutor->>Modal Cloud: Launch sandbox(es) with resources and image
    Modal Cloud-->>ModalSandboxExecutor: Run pipeline/steps and stream logs
    ModalSandboxExecutor->>ModalOrchestrator: Report completion or errors
    ModalOrchestrator->>ZenML: Finalize run status
    ZenML-->>User: Pipeline run results

Suggested reviewers

avishniakov
strickvl

Poem

In Modal clouds our pipelines soar,
With sandboxes spinning, resources galore!
Orchestrator and step, now both take the stage,
With docs and utilities, it’s a brand-new page.
ZenML and Modal, a seamless duet—
Bunnies rejoice, the best run yet! 🐇✨

✨ Finishing touches

🧪 Generate unit tests

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch feature/modal-orchestrator

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

- Adds new Modal orchestrator flavor for serverless pipeline execution - Implements optimized execution modes: pipeline (default) and per_step - Supports GPU/CPU resource configuration with intelligent defaults - Features persistent apps with warm containers for fast execution - Includes comprehensive documentation and examples - Simplifies execution model by removing redundant single_function mode 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>

github-actions · 2025-06-12T09:46:05Z

Documentation Link Check Results

❌ Absolute links check failed
There are broken absolute links in the documentation. See workflow logs for details
✅ Relative links check passed
_{Last checked: 2025-09-12 07:14:32 UTC}

…tepOperator

github-actions · 2025-06-12T11:51:26Z

✅ Branch tenant has been deployed! Access it at: https://staging.cloud.zenml.io/workspaces/feature-modal-orchestrator/projects

strickvl · 2025-06-13T11:43:37Z

src/zenml/integrations/modal/utils.py

+                        elif (
+                            log_age < 300
+                        ):  # Only show logs from last 5 minutes
+                            # This log is recent enough to likely be ours
+                            logger.info(f"{log_msg}")
+                        # Else: skip old logs


This bit seems maybe like it could use more attention. Also are you sure that we wouldn't have time zone mismatches here etc?

strickvl · 2025-06-13T11:44:19Z

src/zenml/integrations/modal/utils.py

+    else:
+        # Fallback to first step's resource settings if no pipeline-level resources
+        if deployment.step_configurations:
+            first_step = list(deployment.step_configurations.values())[0]


what is the first step is unrepresentative, though?

We should check / just make this really explicit in the docs that this is how we assume this

strickvl · 2025-06-13T11:45:43Z

docs/book/component-guide/orchestrators/modal.md

+# Register the orchestrator with explicit credentials
+zenml orchestrator register <ORCHESTRATOR_NAME> \
+    --flavor=modal \
+    --token=<MODAL_TOKEN> \
+    --workspace=<MODAL_WORKSPACE> \
+    --synchronous=true


I think this should use --token-id and --token-secret separately as per the code?

strickvl · 2025-06-13T11:47:59Z

docs/book/component-guide/orchestrators/modal.md

+### Authentication with different environments
+
+For production deployments, you can specify different Modal environments:


Maybe could have a little info box in this section (or maybe even above, linking down here) to say that you might want to have two different stacks, each associated with a different modal environment, one for prod and the other for development etc etc.

docs/book/component-guide/orchestrators/modal.md

strickvl · 2025-06-13T12:12:09Z

src/zenml/integrations/modal/utils.py

+    log_stream_active.set()
+    start_time = time.time()
+
+    def stream_logs() -> None:


Function in function smells a bit wrong and also wondering if we should instead use their Python SDK to stream the logs? https://github.com/modal-labs/modal-client/blob/4177d0b994ac69e01ada7d7a96655c9dcaae570e/modal/cli/utils.py#L24

Possibly something for down the line, though the func-in-func seems off.

strickvl · 2025-06-13T12:13:03Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+if TYPE_CHECKING:
+    from zenml.integrations.modal.flavors.modal_orchestrator_flavor import (
+        ModalOrchestratorConfig,
+        ModalOrchestratorSettings,
+    )
+
+from zenml.integrations.modal.flavors.modal_orchestrator_flavor import (
+    ModalExecutionMode,
+)
+
+if TYPE_CHECKING:
+    from zenml.models import PipelineDeploymentResponse, PipelineRunResponse
+    from zenml.models.v2.core.pipeline_deployment import PipelineDeploymentBase


combine the 'if TYPE_CHECKING' parts?

strickvl · 2025-06-13T12:26:08Z

docs/book/component-guide/orchestrators/modal.md

+
+The Modal orchestrator supports two execution modes:
+
+1. **`pipeline` (default)**: Runs the entire pipeline in a single Modal function for maximum speed and cost efficiency


Not sure I understand why this pipeline option is max speed. Isn't it running everything sequentially in the same container? Wouldn't running things in parallel in separate Modal function calls run faster?

src/zenml/integrations/modal/flavors/modal_orchestrator_flavor.py

strickvl · 2025-06-13T12:30:13Z

docs/book/component-guide/orchestrators/modal.md

+Using the ZenML `modal` integration, you can orchestrate and scale your ML pipelines on [Modal's](https://modal.com/) serverless cloud platform with minimal setup and maximum efficiency.
+
+The Modal orchestrator is designed for speed and cost-effectiveness, running entire pipelines in single serverless functions to minimize cold starts and optimize resource utilization.
+


Maybe some representative screenshot of the Modal UI in here to make the docs a bit friendlier?

i think its fine without

…estrator

- Extract nested log streaming function into ModalLogStreamer class for better code organization - Remove unreliable timezone-based log filtering that could miss logs due to clock skew - Implement smarter resource fallback: use highest requirements across all steps instead of potentially unrepresentative first step - Add logging for resource selection decisions to improve debugging - Fix function-in-function code smell identified in PR review 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>

- Combine duplicate TYPE_CHECKING blocks into single import section - Improve import organization and reduce redundancy - Maintain all existing functionality while improving code structure 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>

- Import MODAL_ORCHESTRATOR_FLAVOR constant from central location to avoid duplication - Update requirements to modal>=1 after testing compatibility with both orchestrator and step operator - Remove unnecessary utils import that was only for mypy discovery - Maintain consistent import patterns across Modal integration files 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>

Based on PR review feedback: - Fix token authentication examples to use --token-id and --token-secret - Add "When NOT to use it" section with clear tradeoffs and alternatives - Add info boxes for environment separation best practices and cost implications - Document Modal vs Step Operator differences with usage recommendations - Add GPU base image requirements and CUDA compatibility warnings - Clarify execution modes: "pipeline" mode reduces overhead vs enables parallelism - Document resource fallback behavior and warming window defaults - Add container warming cost implications with specific guidance - Remove tracking pixel per review request - Improve overall documentation clarity and completeness 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <[email protected]>

schustmi · 2025-07-14T09:10:43Z

src/zenml/integrations/modal/flavors/modal_orchestrator_flavor.py

+        max_parallelism: Maximum number of parallel sandboxes (for PER_STEP mode).
+        synchronous: Wait for completion (True) or fire-and-forget (False).
+    """
+


I think having an option to specify the sandbox name would be nice?

schustmi · 2025-07-14T09:11:59Z

src/zenml/integrations/modal/flavors/modal_orchestrator_flavor.py

+        cloud: The cloud provider to use for the pipeline execution.
+        modal_environment: The Modal environment to use for the pipeline execution.
+        timeout: Maximum execution time in seconds (default 24h).
+        execution_mode: Execution mode - PIPELINE (fastest) or PER_STEP (granular).


Needs more explanation IMO, also the naming I'm not sure about.
Both modes execute your pipeline, but PIPELINE uses a single sandbox and runs sequentially, while PER_STEP spins up a separate sandbox per step.

sandboxes dont have names and the app names are equal to the pipeline names

I think this reply is on the wrong comment. What I meant was actually the app name. I would equate this with specifying a kubernetes namespace or pod name, both of which we allow users to specify but fallback to defaults like the pipeline name if no value is given.

@schustmi ok added

src/zenml/integrations/modal/step_operators/modal_step_operator.py

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

schustmi · 2025-07-14T09:16:51Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        execution_mode = getattr(
+            settings, "execution_mode", ModalExecutionMode.PIPELINE
+        )


Suggested change

execution_mode = getattr(

settings, "execution_mode", ModalExecutionMode.PIPELINE

)

execution_mode = settings.execution_mode

schustmi · 2025-07-14T09:17:29Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+            logger.error(f"Pipeline execution failed: {e}")
+            raise
+
+        logger.info("✅ Pipeline execution completed successfully")


This is only true if the orchestrator is running in sync mode

schustmi · 2025-07-14T09:18:34Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        orchestrator_run_id = str(uuid4())
+        environment[ENV_ZENML_MODAL_ORCHESTRATOR_RUN_ID] = orchestrator_run_id


There is no need to pass this to the entrypoint if it is simply a random UUID. In that case, we can simply generate it in the entrypoint itself.

schustmi · 2025-07-14T09:20:53Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        logger.info(
+            f"🚀 Executing pipeline with Modal ({execution_mode.lower()} mode)"
+        )
+


If the execution mode is pipeline, we do not allow per-step images (as they cannot be used). I suggest you overwrite the get_docker_builds(...) method as follows:

Call the super implementation

Check if there is any build configuration in the result that has key="orchestrator" and a non-None step_name. In that case, raise an error.

…r.py Co-authored-by: Michael Schuster <[email protected]>

Co-authored-by: Michael Schuster <[email protected]>

…l into feature/modal-orchestrator

htahir1

@schustmi Fixed your comments

htahir1 · 2025-07-09T15:52:21Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        )
+
+        # Build Modal image
+        zenml_image = self._build_modal_image(deployment, stack, environment)


Comment: Dont build each time, use metadata of build id to track whether this build is already in modal or not

htahir1 · 2025-07-14T21:12:07Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+            logger.error(f"Pipeline execution failed: {e}")
+            raise
+
+        logger.info("✅ Pipeline execution completed successfully")


htahir1 · 2025-07-14T21:14:36Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        execution_mode = getattr(
+            settings, "execution_mode", ModalExecutionMode.PIPELINE
+        )


htahir1 · 2025-07-14T21:20:07Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        logger.info(
+            f"🚀 Executing pipeline with Modal ({execution_mode.lower()} mode)"
+        )
+


htahir1 · 2025-07-14T21:20:12Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        orchestrator_run_id = str(uuid4())
+        environment[ENV_ZENML_MODAL_ORCHESTRATOR_RUN_ID] = orchestrator_run_id


src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

htahir1 · 2025-07-14T21:22:47Z

src/zenml/integrations/modal/flavors/modal_orchestrator_flavor.py

+        cloud: The cloud provider to use for the pipeline execution.
+        modal_environment: The Modal environment to use for the pipeline execution.
+        timeout: Maximum execution time in seconds (default 24h).
+        execution_mode: Execution mode - PIPELINE (fastest) or PER_STEP (granular).


sandboxes dont have names and the app names are equal to the pipeline names

schustmi · 2025-07-15T11:41:14Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+                    raise ValueError(
+                        f"Per-step Docker settings are not supported in PIPELINE "
+                        f"execution mode. Step '{build.step_name}' has custom Docker "
+                        f"settings but will be ignored since the entire pipeline runs "


It will not be ignored, but instead the run will fail right here with this exception

schustmi · 2025-07-15T11:42:04Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        )
+
+        # Determine if we should wait for completion
+        synchronous = (


synchronous = settings.synchronous

schustmi · 2025-07-15T11:45:24Z

src/zenml/integrations/modal/step_operators/modal_step_operator.py

                        region=settings.region,
                        app=app,
-                        timeout=86400,  # 24h, the max Modal allows
+                        timeout=settings.timeout,


Feel free to ignore as part of this PR as well, but seems like this could also be refactored to use the same code that runs a step of a pipeline in a new sandbox?

schustmi · 2025-07-15T11:48:15Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator_entrypoint.py

+        token_id=orchestrator.config.token_id,
+        token_secret=orchestrator.config.token_secret,
+        workspace=orchestrator.config.workspace,
+        environment=orchestrator.config.modal_environment,


The modal_environment is not a config option and can therefore be overwritten using the settings. So I'm guessing we should probably fetch this from the pipeline settings instead?

schustmi · 2025-07-15T11:48:35Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+            token_id=self.config.token_id,
+            token_secret=self.config.token_secret,
+            workspace=self.config.workspace,
+            environment=self.config.modal_environment,


Same as for the entrypoint, this should be fetched from the settings instead.

schustmi · 2025-07-15T12:13:28Z

src/zenml/integrations/modal/utils.py

+        The configured Modal image.
+    """
+    # Try to get existing image from the app
+    image_name_key = f"zenml_image_{build_id}"


How is this even supposed to work? This key is the same for all images in the build, which are certainly not referring to the same docker image.

schustmi · 2025-07-15T12:14:19Z

src/zenml/integrations/modal/utils.py

+                    logger.warning(f"Invalid memory {memory_mb}MB, ignoring.")
+                    memory_mb = None
+                elif memory_mb < 128:  # Less than 128MB seems too low
+                    logger.warning(


Why would it not be intentional, the user explicitly specified this?

schustmi · 2025-07-15T12:14:46Z

src/zenml/integrations/modal/utils.py

+    )
+
+    # Build Modal image from the ZenML-built image
+    logger.debug(f"Building Modal image from base: {image_name}")


schustmi · 2025-07-15T12:15:12Z

src/zenml/integrations/modal/utils.py

+    zenml_image = (
+        modal.Image.from_registry(
+            image_name, secret=registry_secret
+        ).pip_install("modal")  # Install Modal in the container


modal is installed in the container, as part of the ZenML image build that happened for the stack.

schustmi · 2025-07-15T12:16:25Z

src/zenml/integrations/modal/utils.py

+def _build_modal_image_from_registry(
+    image_name: str,
+    stack: "Stack",
+    environment: Optional[Dict[str, str]] = None,


Didn't we say we don't build these into the image anymore?

If you do, your image reuse doesn't work, as these environment variables contain e.g. credentials to access the ZenML server. If they get reused, they point to potentially a wrong ZenML user, are credentials for the wrong user, ...

schustmi · 2025-07-15T12:18:13Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+        environment: Dict[str, str],
+        settings: ModalOrchestratorSettings,
+        shared_image_cache: Optional[Dict[str, Any]] = None,
+        shared_app: Optional[Any] = None,


schustmi · 2025-07-15T12:21:40Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+            run_id: The pipeline run ID.
+            synchronous: Whether to wait for completion.
+        """
+        logger.debug("Executing entire pipeline in single sandbox")


This is called no matter which mode the orchestrator is running in.

schustmi · 2025-07-15T12:21:51Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+        command = (
+            ModalOrchestratorEntrypointConfiguration.get_entrypoint_command()
+        )
+        from uuid import UUID


Import at the top

schustmi · 2025-07-15T12:22:51Z

src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

+        )
+
+        # Submit the pipeline
+        run_id = str(placeholder_run.id) if placeholder_run else None


You're converting this to a string here, only to convert it back to a UUID in the executor.execute_pipeline method.

schustmi · 2025-07-15T12:23:28Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+                environment_name=settings.modal_environment,
+            )
+
+    def _build_entrypoint_command(


10 lines of code for a +? Can be removed I thjink

schustmi · 2025-07-15T12:35:06Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+        else:
+            return f"{build_id}_pipeline_{image_hash}"
+
+    async def execute_pipeline(


This is more like start_pipeline_sandbox right?

schustmi · 2025-07-15T12:35:45Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+    def _prepare_modal_api_params(
+        self,
+        entrypoint_command: List[str],
+        image: Any,


schustmi · 2025-07-15T12:39:50Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+
+            # Store the image ID for future caching after sandbox creation
+            # The image should be hydrated after being used in sandbox creation
+            await self._store_image_id(zenml_image)


Should probably happen before the sync waiting?

schustmi · 2025-07-15T12:40:52Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+        """
+        try:
+            # After sandbox creation, the image should be hydrated
+            zenml_image.hydrate()


No zenml image has this method, probably because this is a modal image instead. No Any type annotations.

schustmi · 2025-07-15T13:24:25Z

src/zenml/integrations/modal/orchestrators/modal_sandbox_executor.py

+        # Execute pipeline sandbox
+        await self._execute_sandbox(
+            entrypoint_command=entrypoint_command,
+            mode="PIPELINE",


Actually the mode here can be either pipeline or step, because this launches a sandbox that either

runs the entire pipeline

only orchestrator the steps
So we should rather pass the actual mode here, or nothing at all. This is anyway only used for tags.

socket-security · 2025-09-12T07:31:14Z

Review the following changes in direct dependencies. Learn more about Socket for GitHub.

Diff	Package	Supply Chain Security	Vulnerability	Quality	Maintenance	License
	gradio@5.45.0
	transformers@4.28.1
	psutil@7.0.0
	hvac@2.3.0
	numpy@1.22.4
	ipython@8.10.0
	langchain-core@0.3.76
	setuptools@80.9.0
	torch@2.8.0
	transformers@4.56.1
	nltk@3.9.1
	sagemaker@2.251.1
	zenml@0.84.3
	numpy@2.3.3
	litellm@1.77.0
	langchain-community@0.3.29
	pandas@1.5.3
	pandas@2.3.2
	crewai@0.186.1
	scipy@1.16.2
	langchain@0.3.27
	google-cloud-aiplatform@1.112.0
	mypy@1.18.1
	torchvision@0.23.0
	kfp@2.14.3
	strands-agents-tools@0.2.7
	langgraph@0.6.7
	kubernetes@33.1.0
	llama-index@0.14.0
	pytest@8.4.2
	openai-agents@0.3.0
	datasets@2.12.0
	torchaudio@2.8.0
See 96 more rows in the dashboard

View full report

socket-security · 2025-09-12T07:31:16Z

Warning

Review the following alerts detected in dependencies.

According to your organization's Security Policy, it is recommended to resolve "Warn" alerts. Learn more about Socket for GitHub.

Action	Severity	Alert (click "▶" to expand/collapse)
Warn		`[email protected]` has a Critical CVE. CVE: GHSA-3863-2447-669p transformers has a Deserialization of Untrusted Data vulnerability (CRITICAL) Affected versions: < 4.36.0 Patched version: 4.36.0 From: `?` → `pypi/[email protected]` ℹ Read more on: This package \| This alert \| What is a critical CVE? Next steps: Take a moment to review the security alert above. Review the linked package source code to understand the potential risk. Ensure the package is not malicious before proceeding. If you're unsure how to proceed, reach out to your security team or ask the Socket team for help at `[email protected]`. Suggestion: Remove or replace dependencies that include known critical CVEs. Consumers can use dependency overrides or npm audit fix --force to remove vulnerable dependencies. Mark the package as acceptable risk. To ignore this alert only in this pull request, reply with the comment `@SocketSecurity ignore pypi/[email protected]`. You can also ignore all packages with `@SocketSecurity ignore-all`. To ignore an alert for all future pull requests, use Socket's Dashboard to change the triage state of this alert.

View full report

Add Modal orchestrator with step operator and orchestrator flavors

a599f69

htahir1 requested review from strickvl and safoinme June 12, 2025 09:36

github-actions bot added internal To filter out internal PRs and issues enhancement New feature or request labels Jun 12, 2025

htahir1 added 3 commits June 12, 2025 12:04

Add pipeline-wide resource settings for hardware resources

cd6b59f

Add Modal step operator orchestrator to run pipelines on Modal

58d72c5

Add log streaming for async execution in Modal Orchestrator

9a0f2a5

htahir1 added the staging-workspace label Jun 12, 2025

htahir1 added 2 commits June 12, 2025 12:33

Add app warming window hours for container reuse

7cf8fd5

Remove unnecessary exception handling in ModalOrchestrator and ModalS…

60d227a

…tepOperator

htahir1 added 7 commits June 12, 2025 13:54

Refactor return statement to assign and return validator

45ed008

Add Modal integration utils module for mypy discoverability

0b6b9f9

Update Modal integration to require Modal version 1

d8318b1

Introduce ModalExecutionMode enum for execution modes

b7400b6

Refactor Modal authentication setup and deployment

ee15117

Update platform references in log messages

85aeef4

Include pipeline run ID for isolation and conflict prevention

01f7ef2

strickvl reviewed Jun 13, 2025

View reviewed changes

htahir1 and others added 5 commits June 13, 2025 14:35

Merge remote-tracking branch 'origin/develop' into feature/modal-orch…

6deba69

…estrator

htahir1 requested a review from strickvl June 13, 2025 14:25

Refactor error message formatting and GPU count handling

f2218b1

htahir1 added 2 commits July 13, 2025 21:24

Add Modal orchestrator integration tests

083320f

Update logging messages to be more descriptive

15e27a8

schustmi requested changes Jul 14, 2025

View reviewed changes

htahir1 and others added 6 commits July 14, 2025 14:14

Merge branch 'develop' into feature/modal-orchestrator

4f4ab4a

Update src/zenml/integrations/modal/step_operators/modal_step_operato…

5f4c69a

…r.py Co-authored-by: Michael Schuster <[email protected]>

Update src/zenml/integrations/modal/orchestrators/modal_orchestrator.py

bbbe9df

Co-authored-by: Michael Schuster <[email protected]>

Update execution mode setting to "mode" in Modal orchestrator

0208989

Refactor Modal orchestrator code for better readability

83d1bbc

Merge branch 'feature/modal-orchestrator' of github.com:zenml-io/zenm…

b7d8cc2

…l into feature/modal-orchestrator

htahir1 commented Jul 14, 2025

View reviewed changes

htahir1 requested a review from schustmi July 14, 2025 21:23

htahir1 added 7 commits July 14, 2025 23:26

Add Docker configuration validation for Modal orchestrator

bb9790b

Refactor variable names for execution mode in ModalOrchestrator

5b09c58

Refactor log statements for better readability

2e618d7

Update ModalOrchestrator to submit pipelines to Modal

44121d1

Refactor method signature in ModalOrchestrator class

099264e

Add custom app name option for Modal orchestrator

8d6bfec

Add pydantic Field descriptions to ModalStepOperatorSettings

8ec1c08

schustmi requested changes Jul 15, 2025

View reviewed changes

schustmi and others added 7 commits July 29, 2025 14:25

Start some cleanup

d272580

Merge branch 'develop' into feature/modal-orchestrator

0160c16

Next round of cleanup

1490d39

More cleanup

6f23076

Use correct key

0533723

Add todo

a638328

Merge branch 'develop' into feature/modal-orchestrator

f15ce9c

strickvl mentioned this pull request Sep 29, 2025

Add Modal deployer flavor #4012

Open

1 task

		### Authentication with different environments

		For production deployments, you can specify different Modal environments:


		The Modal orchestrator supports two execution modes:

		1. `pipeline` (default): Runs the entire pipeline in a single Modal function for maximum speed and cost efficiency

		Using the ZenML `modal` integration, you can orchestrate and scale your ML pipelines on [Modal's](https://modal.com/) serverless cloud platform with minimal setup and maximum efficiency.

		The Modal orchestrator is designed for speed and cost-effectiveness, running entire pipelines in single serverless functions to minimize cold starts and optimize resource utilization.

		orchestrator_run_id = str(uuid4())
		environment[ENV_ZENML_MODAL_ORCHESTRATOR_RUN_ID] = orchestrator_run_id

Add Modal orchestrator with step operator and orchestrator flavors #3733

Are you sure you want to change the base?

Add Modal orchestrator with step operator and orchestrator flavors #3733

Uh oh!

Conversation

htahir1 commented Jun 12, 2025 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pre-requisites

Types of changes

Summary by CodeRabbit

Uh oh!

coderabbitai bot commented Jun 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Walkthrough

Changes

Sequence Diagram(s)

Suggested reviewers

Poem

Uh oh!

github-actions bot commented Jun 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Documentation Link Check Results

Uh oh!

github-actions bot commented Jun 12, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

htahir1 left a comment

htahir1 commented Jun 12, 2025 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Jun 12, 2025 •

edited

Loading

github-actions bot commented Jun 12, 2025 •

edited

Loading