| CREATE CONNECTION | Creates a new data source from an external connection. |
| CREATE DATASET | Creates a new dataset from a source in the current connection. |
| CREATE DEPLOYMENT | Creates or updates a deployment for a trained model. |
| CREATE MODEL | Used to define and train a new Ludwig model configuration. |
| CREATE SCHEDULE | Allows a SQL or PQL statement to be executed on a schedule. |
| DROP CONNECTION | Deletes an existing connection and any dependant datasets and models. |
| DROP DATASET | Deletes an existing dataset and any dependant models. |
| DROP DEPLOYMENT | Deletes an existing deployment. |
| DROP MODEL | Deletes an existing model repository. |
| EVALUATE | Used to evaluate one or more target features from a trained model given a query input. |
| EXPLANATION | Using PREDICT, you have the option to add the clause WITH EXPLANATION to generate explanations for the model's predictions. |
| EXPORT MODEL | Exports the model to current object store connection to an export location with optional framework. |
| FORK MODEL | Used to create a new model that is a copy an existing model. |
| PREDICT | Used to compute predictions for one or more target features from a trained model given an input selection. |
| PROMPT | Used to query large language models by providing a prompt and optional input selection. |
| RETRAIN MODEL | Used to retrain an existing model with the latest dataset. |
| SELECT | Used to obtain rows from the dataset and to filter results returned by prediction queries. |
| SHOW DEPLOYMENTS | Returns a list of active deployments, optional filtering by serving engine, repository or name. |