AI module

What is the AI module for?

The AI module provides a unified interface for accessing various AI services from different providers. The SAWS connector and the OWL Data Hub offer to use AI functionality for vectorization, translation and other large language model capabilities. Different providers can be used for these services as a backend.

Navigating to the AI module

The AI module can be found on the dashboard. Here you can access the AI API configuration for provider-specific settings and the AI Assignment configuration for purpose-specific settings. 

If the tile is shown as deactivated, you do not have a license. In this case, please contact sawsconnector@saws.de.

If you are not on the dashboard, click the leftmost button in the toolbar to navigate to the dashboard page.

AI API Config

In order to use the AI functionality, you need to request an API key from your chosen provider and store it here. You can create a new API configuration by clicking the plus button in the top-left corner of the table.

This will open an editor with the following settings:

• Name
• Provider
• API key
• Description

Fields

Name

The name should be descriptive and unique.

Provider

The provider corresponding to the API key.

API key

This is where you can store your API key. You will need to provide your own API key from the configured provider.

Provider specific settings

Some providers require some additional settings. For example, an API URL is sometimes required. These fields will only appear if they are needed.

Description

You may provide a description for your own use.

Assignments

The final row shows where your API configuration is currently being used.

Once everything is configured you may save the API configuration and return to the previous page. 

Overview

Once you have configured everything, you can save the API configuration and return to the previous page. You can delete unused keys or edit them, which will take you back to the edit page.

Note

If the “Provider” field is greyed out when editing an existing AI API configuration, this is because it is configured in an AI Assignment configuration. Check the “Assignments” field to see where the AI API configuration is being used.

AI assignment

The assignment allows you to set specific settings for various AI purposes. Navigation in the AI Assignment view is very similar to the AI API Config.  You can create a new Assignment configuration by clicking the plus button in the top left corner of the table.

This will open an editor with the following settings:

• Name
• Transformation
• API key config
• Model
• Default switch
• Description

Fields

Name

The name should be descriptive and unique.

Purpose

The purpose describes the AI functionality that should be used. Options are Master, Transformation, Translation, Assignment, Population, Image, Vectorization.

API key config

The API key configuration is one of the API keys that were configured in the previous view.

Model

Most providers offer a range of models. Once you have selected the purpose and API key configuration, suggestions for available models will be shown. If you want to use a different model, make sure that it is actually available with your provider and API key.

Default

For every purpose there is one default assignment configuration that is used if no specific assignment configuration is specified.

Description

You may provide a description for your own use.

Vectors

The last row will show where your Assignment configuration is currently being used.

Once everything is configured you may save the Assignment configuration and return to the previous page. 

Overview

Once you return to the main AI Assignment page, you will see an overview of all your Assignment configs. You may delete unused configs, edit configs, which will bring you back to the edit page. The assignment overview may show some preset configs for each purpose. Those are the recommended settings for every given purpose. They will only show if a fitting AI API key is configured. Note that the presets may change if new API keys are entered, or the recommendations are changed.

Note

If some of the fields are greyed out when editing an existing AI Assignment configuration, this is because it is actively used. Check the Vectors field to see where the AI Assignment configuration is being used.

Was this article helpful?

0 0

How can we improve this article?

Please submit the reason for your vote so that we can improve the article.

Name

Email

Submit