Delete an MLflow run by moving it to the 'deleted' lifecycle stage. The run is hidden from UI and queries but remains recoverable through the MLflow API.
MIT
458,267 tools. Updated 2026-08-15 17:54
"MLflow" matching MCP tools:
- Verify MLflow server status and connectivity to ensure the server is reachable and operational.MIT
- Retrieve attachments linked to a trace by supplying its ID; used to inspect trace metadata in Databricks MLflow.MIT
- Retrieve a trace attachment using trace ID and attachment ID. Compatible with Databricks MLflow (OSS servers return 404).MIT
- Delete an MLflow experiment by providing its experiment ID. Removes the experiment and associated runs from the Databricks workspace.MIT
- Retrieve an MLflow experiment's details using its experiment ID for analysis or monitoring.MIT
Matching MCP Servers
- AlicenseBqualityAmaintenanceMLflow MCP — 82 tools for experiments, runs, registered models, traces, assessments. MLflow 3 GenAI traces support with analyze-failed-traces workflow Prompt.82851MIT
- AlicenseAqualityBmaintenanceA Model Context Protocol server that enables LLMs to interact with MLflow tracking servers, allowing users to query experiments, analyze runs, compare metrics, manage the model registry, and promote models through natural language.4014MIT
- Delete a specific version of a registered MLflow model by providing the model name and version number.MIT
- Retrieve a specific model version from a registered MLflow model by providing the model name and version number.MIT
- Create a new MLflow experiment to track machine learning runs with unique name, optional artifact location, and tags.MIT
- Create a logged model entity in an MLflow experiment by specifying experiment ID, model name, type, source run, parameters, and tags.MIT
- Search and list MLflow experiments with optional filters, pagination, and sorting to manage experiment tracking in Databricks.MIT
- Create a new version of a registered MLflow model by specifying the model name and source URI.MIT
- Update an MLflow experiment's name, tags, or lifecycle stage using the Databricks REST API.MIT
- Create a new MLflow experiment with a required name and optional artifact location and tags.MIT
- List and filter MLflow model versions using search criteria such as filter, ordering, and pagination.MIT
- List all Managed MLflow Tracking Servers in SageMaker to review and manage your MLflow environments.Apache 2.0
- Execute bash commands within the sktime container to install Python packages or inspect files.BSD 3-Clause
- Save an estimator or pipeline handle to a specified path using sktime MLflow integration for model persistence and management.BSD 3-Clause
- Search for logged models across one or more experiments using SQL-like filters, dataset conditions, and sorting to find specific models.MIT
- Transition a model version to a new stage (Staging, Production, or Archived), with an option to archive existing versions in the target stage.MIT