Apply vector embeddings to user queries with the OpenAI embeddings generator

The OpenAI embeddings generator user experience tuning stage allows you to connect to your OpenAI embeddings model to automatically transform user queries to apply vector embeddings to the query during runtime.

Prerequisites

Note the following prerequisites before configuring the AI Visual Interpreter component:

  • If you are using OpenAI, the model you use must be gpt-4o or later.

To configure your OpenAI embeddings generator, do the following:

  1. In the SmartHub administration portal, click General Settings.

  2. In the User Experience Tuning section, click ADD QUERY TUNING or ADD RESULTS TUNING.

  3. In the User Experience Tuning field, select OpenAI embeddings generator. Various new fields will appear on the right side of the User Experience Tuning window.

  4. In the Name field, enter a name for your tuning stage.

  5. In the Model Name field, enter the model name for your Open AI model. For more information, see the Open AI documentation on Embedding models.

  6. In the API Key field, enter the API key for your Azure Open AI instance.

  7. In the Output Metadata Name field, enter a name for the metadata that your vectors embeddings will be generated under. For example, BodyEmbeddings.

  8. Click Validate Connection and confirm that you are successful in connecting to your API endpoint.

  9. Click Accept.