Support uv in mlflow.models.predict

[serena-ruan]

🛠 DevTools 🛠

Install mlflow from this PR

pip install git+https://github.com/mlflow/mlflow.git@refs/pull/13824/merge

Checkout with GitHub CLI

gh pr checkout 13824

Related Issues/PRs

#xxx

What changes are proposed in this pull request?

Support uv as a env manager for running mlflow.models.predict

How is this PR tested?

• Existing unit/integration tests
• New unit/integration tests
• Manual tests

Does this PR require documentation update?

• No. You can skip the rest of this section.
• Yes. I've updated:
  • Examples
  • API references
  • Instructions

Release Notes

Is this a user-facing change?

• No. You can skip the rest of this section.
• Yes. Give a description of this change to be included in the release notes for MLflow users.

What component(s), interfaces, languages, and integrations does this PR affect?

Components

• area/artifacts: Artifact stores and artifact logging
• area/build: Build and test infrastructure for MLflow
• area/deployments: MLflow Deployments client APIs, server, and third-party Deployments integrations
• area/docs: MLflow documentation pages
• area/examples: Example code
• area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
• area/models: MLmodel format, model serialization/deserialization, flavors
• area/recipes: Recipes, Recipe APIs, Recipe configs, Recipe Templates
• area/projects: MLproject format, project running backends
• area/scoring: MLflow Model server, model deployment tools, Spark UDFs
• area/server-infra: MLflow Tracking server backend
• area/tracking: Tracking Service, tracking client APIs, autologging

Interface

• area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
• area/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Models
• area/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registry
• area/windows: Windows support

Language

• language/r: R APIs and clients
• language/java: Java APIs and clients
• language/new: Proposals for new client languages

Integrations

• integrations/azure: Azure and Azure ML integrations
• integrations/sagemaker: SageMaker integrations
• integrations/databricks: Databricks integrations

How should the PR be classified in the release notes? Choose one:

• rn/none - No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" section
• rn/breaking-change - The PR will be mentioned in the "Breaking Changes" section
• rn/feature - A new user-facing feature worth mentioning in the release notes
• rn/bug-fix - A user-facing bug fix worth mentioning in the release notes
• rn/documentation - A user-facing documentation change worth mentioning in the release notes

Should this PR be included in the next patch release?

Yes should be selected for bug fixes, documentation updates, and other small changes. No should be selected for new features and larger changes. If you're unsure about the release classification of this PR, leave this unchecked to let the maintainers decide.

What is a minor/patch release?

• Minor release: a release that increments the second part of the version number (e.g., 1.2.0 -> 1.3.0).
  Bug fixes, doc updates and new features usually go into minor releases.
• Patch release: a release that increments the third part of the version number (e.g., 1.2.0 -> 1.2.1).
  Bug fixes and doc updates usually go into patch releases.

• Yes (this PR will be cherry-picked and included in the next patch release)
• No (this PR will be included in the next minor release)

[github-actions]

Documentation preview for 4e53187 will be available when this CircleCI job
completes successfully.

• Top page
• Changed pages

More info

• Ignore this comment if this PR does not change the documentation.
• It takes a few minutes for the preview to be available.
• The preview is updated when a new commit is pushed to this PR.
• This comment was created by https://github.com/mlflow/mlflow/actions/runs/12234139821.

[harupy]

Imagine you manage everything in uv and have a uv.lock file. In that situation, you would want to log uv.lock file as an artifact and use it to restore the environment. Should we support it or not?

[BenWilson2]

Can we update https://mlflow.org/docs/latest/model/dependencies.html#how-to-fix-dependency-errors-when-serving-my-model section to fully explain and provide a comparison for package managers (uv >> pip >> conda for speed) and to show overriding the env manager option when calling the API? We want to gently push people to using this.

Let's also make sure to update https://github.com/mlflow/mlflow/blob/master/mlflow/models/python_api.py#L111-L178 the code block here to show overriding the env_manager parameter to 'uv'. The API docstring could benefit from an explanation of why we're offering this as an option (otherwise users will just gloss over this and not see the benefit).

We should probably mention this API where it is most useful too in the docs - in a section right above here: https://mlflow.org/docs/latest/model/signatures.html#id1 as this main page for signatures is frequently visited and this API should be featured here with a brief explanation (and links to the more verbose update section above in dependency error fixing).

[serena-ruan]

Can we update https://mlflow.org/docs/latest/model/dependencies.html#how-to-fix-dependency-errors-when-serving-my-model section to fully explain and provide a comparison for package managers (uv >> pip >> conda for speed) and to show overriding the env manager option when calling the API? We want to gently push people to using this.

Let's also make sure to update https://github.com/mlflow/mlflow/blob/master/mlflow/models/python_api.py#L111-L178 the code block here to show overriding the env_manager parameter to 'uv'. The API docstring could benefit from an explanation of why we're offering this as an option (otherwise users will just gloss over this and not see the benefit).

We should probably mention this API where it is most useful too in the docs - in a section right above here: https://mlflow.org/docs/latest/model/signatures.html#id1 as this main page for signatures is frequently visited and this API should be featured here with a brief explanation (and links to the more verbose update section above in dependency error fixing).

Make sense, I'll file a follow-up doc PR for this

[serena-ruan]

@harupy Do we have other concerns about this PR?

[BenWilson2]

Can we add a validation test with a lightweight package where we're overriding a package requirement for a model and verify that the updated version is installed without issue (custom pyfunc that just calls __version__ on the update package as a return value is probably sufficient).
It would be nice to ensure that we're testing both virtualenv and uv functionality for this to make sure that both handle this properly (this should just work for virtualenv, but my only concern is that uv might have some secret caching in place that might behave differently than we expect).

[serena-ruan]

mlflow/tests/models/test_python_api.py

Lines 88 to 92 in a28bf5e

	@pytest.mark.parametrize(
	"env_manager",
	[VIRTUALENV, CONDA, UV],
	)
	def test_predict_with_pip_requirements_override(env_manager):

is the test :)

[BenWilson2]

LGTM once https://github.com/mlflow/mlflow/pull/13824/files#r1871739656 is addressed! Thanks for this feature - it's going to be great :D

[harupy]

Left a few more comments, otherwise LGTM!