Setting up storage and uploading the model

To deploy a custom foundation model for inferencing with watsonx.ai, you must prepare a properly-sized Persistent Volume Claim on your cluster, upload the model, and then make the required conversions. Detailed steps depend on where your model is located.

Service The required watsonx.ai service and other supplemental services are not available by default. An administrator must install these services on the IBM Cloud Pak for Data platform. To determine whether a service is installed, open the Services catalog and check whether the service is enabled.

You must be a Cloud Pak for Data administrator to upload the custom model. Follow the prerequisites for all custom foundation models, then choose the steps for your deployment scenario.

Prerequisites

Public model repositories might require you to set up an account. For example, if you decide to get your model from Hugging Face, you must have a Hugging Face account. To create a new account, go to the Hugging Face website. After creating a new account, generate a new Hugging Face token. To generate a token, see Hugging Face's guide to creating a token.

Set up storage and upload a model. Find a scenario that matches your model:

Next steps

Registering a custom foundation model

Parent topic: Deploying a custom foundation model