Train or Upload a Model

Rasa X can act as a model server. To use it as a model server, you need to add a model to the model's screen. To iterate efficiently, you should automate training, uploading, and tagging a model as part of a CI/CD pipeline.

Adding a Model#

You can add a model to Rasa X in the following ways:

1. Using the HTTP API
2. Using rasactl
3. Using the UI to upload a model from your file system.
4. By training a model from within Rasa X. Hitting the train button within Rasa X will send a training request including the data and config currently in Rasa X to the Rasa Open Source Server that is connected as the training/worker environment. Training in the UI should only be done during development; production models should always be uploaded from a CI/CD pipeline.

Deploying a Model#

To deploy your trained model, you need to tag it as active (Rasa X) or production (Rasa Enterprise). Once you have a model available on the Models screen, you can either tag the model using the UI or use rasactl.