Create a new MLflow experiment

Search experiments with filter and pagination

Get experiment details by ID

Get experiment details by name

Rename an experiment

Soft-delete an experiment by ID

Restore a deleted experiment

Set a tag on an experiment

Delete a tag from an experiment

Create a new run in an experiment

Get run details by ID

Search runs with filter expression and pagination

Update run status, end time, or name

Soft-delete a run by ID

Restore a deleted run

Log a single metric value to a run

Log a single parameter to a run

Log a batch of metrics, params, and tags to a run

Log dataset inputs to a run

Get full history of a metric for a run

Set a tag on a run

Delete a tag from a run

List artifacts under a run's artifact directory

Find the run with the best (max/min) value of a metric in an experiment

Side-by-side metric/param comparison across multiple runs. Renders an Apps SDK card on ChatGPT clients (Claude clients receive the same JSON text).

Find runs whose tags match all of the given key/value pairs

Create a new registered model in the model registry

Get a registered model by name

Search registered models

Rename a registered model

Update a registered model's description

Delete a registered model and all its versions

Get the latest model versions per stage

Set a tag on a registered model

Delete a tag from a registered model

Set an alias on a registered model version

Delete an alias from a registered model

Get the model version pointed to by an alias

Create a new model version

Get a model version by name and version

Search model versions with filter and pagination

Update a model version's description

Delete a specific model version

Transition a model version to a new stage

Get the artifact download URI for a model version

Set a tag on a model version

Delete a tag from a model version

Create a new MLflow 3 LoggedModel entity in an experiment

Search LoggedModels by experiment with filter and pagination

Get a LoggedModel by ID

Set a terminal status (READY/FAILED/...) on a LoggedModel

Soft-delete a LoggedModel by ID

Set or upsert tags on a LoggedModel

Delete a tag from a LoggedModel

Log parameters on a LoggedModel

Search and filter traces in experiments

Retrieve detailed trace information by trace ID

Retrieve trace metadata only (no spans)

Delete traces by ID or older than a timestamp

Add a custom key-value tag to a trace

Remove a tag from a trace

List attachments on a trace (Databricks MLflow only — OSS servers return 404)

Get a specific attachment on a trace by ID (Databricks MLflow only — OSS servers return 404)

Log evaluation feedback (score or judgment) on a trace

Log a ground-truth expectation on a trace

Get an assessment by trace ID and assessment ID

Update an existing assessment

Delete an assessment from a trace

Register a webhook for model registry events

List webhooks (optionally filtered by model name)

Get webhook details by ID

Update an existing webhook (events, URL, status, secret)

Delete a webhook by ID

Send a test event to a webhook to verify configuration

Create a prompt optimization job to automatically improve a registered prompt

Get a prompt optimization job by ID

Search prompt optimization jobs in an experiment

Cancel a running prompt optimization job

Delete a prompt optimization job

Aggregated run view: run info + (optional) metric history + (optional) artifacts list in a single call. Replaces 3-4 round-trips of get-run + get-metric-history (per metric) + list-artifacts.

Aggregated experiment view: experiment overview + topN runs (sorted by metric or start_time) + metric stats (min/max/mean across topN) in a single call. Replaces 3-5 round-trips of get-experiment + search-runs + get-best-run.

Discover available tools by natural language query. Returns matching tool names + descriptions across all categories. Use this first to navigate the 77+ tool surface efficiently.