MLflow MCP Server
About
Integrates with MLflow, enabling AI assistants to interact with experiments, runs, and registered models.
Details
- Author
- yesid-lopez
- Categories
- Developer Tools, AI, Other
Jump to
Setup
Install MLflow MCP Server in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/yesid-lopez/mlflow-mcp-server
Follow the installation instructions in the repository README, then restart your MCP client.
Integrates with MLflow, enabling AI assistants to interact with experiments, runs, and registered models.
- Quickstart
- Tools
- Example Prompts
- Configuration
- Installation (Development)
- Project Structure
- License
The fastest way to get started is to add the server to your MCP client config. No local clone required.
Add to~/Library/Application Support/Claude/claude_desktop_config.json(macOS) or%APPDATA%\Claude\claude_desktop_config.json(Windows):
{ "mcpServers": { "mlflow": { "command": "uvx", "args": ["mlflow-mcp-server"], "env": { "MLFLOW_TRACKING_URI": "http://localhost:5000" } } } }
{ "mcpServers": { "mlflow": { "command": "uvx", "args": ["mlflow-mcp-server"], "env": { "MLFLOW_TRACKING_URI": "http://localhost:5000" } } } }
{ "$schema": "https://opencode.ai/config.json", "mcp": { "mlflow": { "type": "local", "command": ["uvx", "mlflow-mcp-server"], "environment": { "MLFLOW_TRACKING_URI": "http://localhost:5000" } } } }
Replacehttp://localhost:5000with the URL of your MLflow tracking server.
Once configured, you can ask your AI assistant things like:
- "List all experiments related to recommendation models"
- "Show me the runs for experiment 12 and compare their metrics"
- "Get the parameters and metrics for run abc123"
- "Which runs in the fraud-detection experiment have the highest accuracy?"
- "Show me all registered models"
- "Register a new model called churn-classifier with description 'Binary classifier for customer churn'"
- "Create a new version of churn-classifier from run abc123"
- "Set the champion alias on version 3 of churn-classifier"
- "Rename the model old-name to new-name"
- "Delete version 1 of churn-classifier"
- "Compare the last 5 runs of the search-ranking experiment by NDCG and latency"
- "What hyperparameters were used in the best-performing run of experiment 7?"
- "List all model versions for recommendation-model and their aliases"
- Python 3.11+
- uv
- An MLflow tracking server
git clone https://github.com/yesid-lopez/mlflow-mcp-server.git cd mlflow-mcp-server uv sync
export MLFLOW_TRACKING_URI="http://localhost:5000" uv run -m mlflow_mcp_server
The server communicates overstdio, which is the standard MCP transport for local tool servers.
mlflow_mcp_server/ ├── __main__.py # Entry point ├── server.py # MCP server setup and tool registration ├── tools/ │ ├── experiment_tools.py # Experiment search and retrieval │ ├── run_tools.py # Run details and listing │ └── registered_models.py # Model registry CRUD operations └── utils/ └── mlflow_client.py # MLflow client singleton
- Create a function in the appropriate file undertools/.
- Register it inserver.py:
from mlflow_mcp_server.tools.your_module import your_function mcp.add_tool(your_function)
uv run ruff check . uv run ruff format --check .
This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.
Solve optimization problems formulated by an LLM using the on-device Gurobi solver.
Train a Linear Regression model by uploading a CSV dataset file, demonstrating an end-to-end machine learning workflow.
Neo is the first autonomous AI engineering agent that plans, researches and executes multi-step work for tasks such as building AI agents, AI model fine-tuning and evals, and ML pipelines; using your codebase, data, and experiments to ship faster with inspectable artifacts. It can reason over your repository, wire tools and retrieval, debug training runs, and help you develop production-ready AI workflows
A reasoning-first MCP middleware that uses heuristics and Neural BERT models to distil context and eliminate noise
An MCP server for accessing tidymodels GitHub information and generating code.
The MCP server provides shape of tensors to convert PyTorch code to einsum and einops
MCP server that gives AI assistants on-demand access to 1,500+ amCharts docs, ~300 code examples, and 1000+ class API references.
A Model Context Protocol server for generating visual charts using AntV.
Query and analyze your Opik logs, traces, prompts and all other telemtry data from your LLMs in natural language.
Get access to Kaggle's datasets, models, competitions, notebook and benchmarks.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
