mlflow-mcp-server MCP Server

A Model Context Protocol (MCP) server that exposes MLflow experiment tracking and model registry operations as tools for AI assistants.

Local serverstdio

What is the mlflow-mcp-server MCP Server?

mlflow-mcp-server is an MCP project maintained in the yesid-lopez/mlflow-mcp-server repository. The maintainer describes its purpose as “A Model Context Protocol (MCP) server that exposes MLflow experiment tracking and model registry operations as tools for AI assistants”. Read the project documentation for the complete scope.

Installation

The documented pypi package is mlflow-mcp-server. The configuration below comes from the maintainer’s example; adapt its paths and account values to your environment.

Configuration values

The example declares MLFLOW_TRACKING_URI. Secret-looking values have been replaced with placeholders. Required and optional labels are shown only when the documentation states them.

Source

Project documentation · Repository · Discovery listing. This profile records the public documentation checked on 2026-10-08.

Recorded tools

Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.

Tool	What it does
get_experiment	Get experiment details by ID
get_experiment_by_name	Get experiment details by name
search_experiments	List and filter experiments with optional name matching and pagination
get_run	Get full run details including metrics, parameters, tags, and run type (parent/child/standalone)
get_experiment_runs	List runs for an experiment with pagination
get_registered_models	Search and list registered models
get_model_versions	Browse model versions with filtering
create_registered_model	Create a new registered model with optional description and tags
create_model_version	Create a new model version from a run's artifacts
rename_registered_model	Rename an existing registered model
set_registered_model_alias	Assign an alias (e.g. champion, challenger) to a model version
delete_registered_model	Delete a registered model and all its versions
delete_model_version	Delete a specific model version

How to install the mlflow-mcp-server MCP Server

{
  "mcpServers": {
    "mlflow": {
      "command": "uvx",
      "args": [
        "mlflow-mcp-server"
      ],
      "env": {
        "MLFLOW_TRACKING_URI": "http://localhost:5000"
      }
    }
  }
}

From the project documentation. Replace placeholders and example paths before connecting.