GenAI Icon

Unqork GenAI Connector

Overview

Unqork's GenAI Connector enables Creators Also known as Unqork Users, or Designer Users; is anyone who is inside the Unqork platform. to add GenAI Generative AI (GenAI) is a type of artificial intelligence (AI) that can create new content in response to written prompts. capabilities to Unqork applications. To use the GenAI Connector, Creators install the template to their environment, then connect it to the OpenAI or Google AI API APIs (application programming interfaces) are a set of protocols and definitions developers use to build and integrate application software. APIs act as the connective tissue between products and services. services. Creators can then customize LLM Large Language Models (LLM) are artificial intelligence (AI) programs that use deep learning algorithms to analyze and generate text and other content based on large amounts of data. usage using the GenAI Connector's Plug-In components. Once the connector is set up, Creators and end-users End-users, also known as Express Users, are the individuals accessing an application through Express View. In most cases, end-users are the customers using the product. can interact with the LLM in Express View Express View is how your end-user views your application. Express View also lets you preview your applications to test your configuration and view the styling. This is also the view your end-users will see when interacting with your application. After configuring a module, click Preview in the Module Builder to interact with the module in Express View..

The GenAI Connector is available for the UDesigner platform.

What Is the GenAI Connector?

The GenAI Connector is a template that provides a starting point for creating GenAI enabled applications. The template contains a sample module configuration for a chatbot A chatbot is a computer program that simulates conversation with end-users. experience that can be customized to suit your needs.

The GenAI Connector package includes the following:

Creators are responsible for acquiring an API key for the LLM provider of their choice. Never share your API key with outside sources.

Installing and Configuring the GenAI Connector

To use an LLM in Express View, you must first install and configure the GenAI Connector.

Requirements

To install and configure GenAI in Unqork, you need:

  • Access to the UDesigner Platform.

  • Subscription to OpenAI's ChatGPT or Google's Gemini AI.

  • Environment Administrator Access.

Install the GenAI Connector Template

The GenAI Connector template creates all the modules and services required to use ChatGPT or Gemini in Unqork. After installing the template, you'll connect it to your preferred service using your private LLM Large Language Models (LLM) are artificial intelligence (AI) programs that use deep learning algorithms to analyze and generate text and other content based on large amounts of data. API key.

To install the GenAI Connector template:

1. From the UDesigner Homepage, navigate to the Unqork and AI card.
2. Click Get Started.

A static image displaying the UDesigner Homepage. The Unoqrk and AI card in the discover bar is highlighted.

3. In the confirmation modal A modal is a window that appears on top of the content you are currently viewing., click Create GenAI Connector App. The Create GenAI Connector App modal displays.
4. From the Workspace* drop-down, enter or select an existing workspace.
5. In the App Name* field, enter a name for your GenAI Connector application. The App Path* populates using the App Name field value.

The App Type*, App Entrypoint* and Style values are predefined and cannot be changed at this time.

6. Click Create App from Template

A static image displaying the Create GenAI Connector App modal.

The GenAI Connector app displays in the Application Element table of the chosen workspace.

Configuring GenAI Connector Services

The GenAI Connector comes with two preconfigured services, Google's Gemini and OpenAPI's ChatGPT. You can connect one, or both services to the GenAI Connector application.

Connect the OpenAI Service

Now that the GenAI Connector template is installed to the environment, connect it to the OpenAI service using the appropriate API key.

For this example, connect the OpenAI API to Unqork using the Services Administration page:

1. From the UDesigner platform, click Administration.
2. Under Integration, click Integration Administration. The Integration Gateway page displays.

If you do not have Integration Administration enabled in your environment, click Services Administration.

3. From the left menu, click Services Administration.

In the Services list, the GenAI Connector template automatically creates the Google Gemini and OpenAI connectors.

4. From the Services list, click OpenAI - GenAI Connector 1.0.0 20XX-XX-XXT00:00:00.000Z. The date and Zulu Zulu Time (also known as Coordinated Universal Time) is an established universal time based on GMT (Greenwhich Mean Time). time correspond to the GenAI Connector template installation date.

A static image displaying the environment's services list.

The OpenAI - GenAI Connector's Service page displays.

5. At the top right, click  Edit. The service fields become available to edit.

If multiple GenAIs connectors exist in the environment, update the Service Title* field to a unique value for easier navigation.

6. From the left menu, click Service Type.
7. In the Bearer Token field, enter your Open AI API access key.

To obtain an API key, view OpenAI's documentation here: https://platform.openai.com/docs/api-reference/authentication

A static image displaying the OpenAI Connector's Service page. The Bearer token field is highlighted.

8. Click Save Changes.

The GenAI Connector's OpenAI application is now ready to use.

Connect the Google Gemini API Service

Now that the GenAI Connector template is installed to the environment, connect it to the Gemini AI service using the appropriate API key.

For this example, connect the Google Gemini API to Unqork using the Services Administration page:

1. From the UDesigner platform, click Administration.
2. Under Integration, click Integration Administration. The Integration Gateway page displays.

If you do not have Integration Administration enabled in your environment, click Services Administration.

3. From the left menu, click Services Administration.

In the Services list, the GenAI Connector template automatically creates the Google Gemini and OpenAI connectors.

4. From the Services list, click Google Gemini - GenAI Connector 1.01.0.0 20XX-XX-XXT00:00:00.000Z. The date and Zulu Zulu Time (also known as Coordinated Universal Time) is an established universal time based on GMT (Greenwhich Mean Time). time correspond to the GenAI Connector template installation date.

A static image displaying the environment's services list.

The Google Gemini connector's Service page displays.

5. In the UDesigner's top-right corner, click  Edit. The service fields become available to edit.

If multiple GenAIs connectors exist in the environment, update the Service Title* field to a unique value for easier navigation.

6. From the left menu, click Request & Response.
7. In the Request Headers table, replace the Header Value of {{YOUR_API_KEY_HERE}} with your Google Gemini API key.

To obtain an API key, view Google's documentation here: https://ai.google.dev/gemini-api/docs/api-key

A static image displaying the Google Connector's Service page. The Request and Response Header is highlighted.

8. Click Save Changes.

The GenAI Connector's Google Gemini application is now ready to use.

Customizing the GenAI Connector's LLM

ChatGPT and Gemini use modifiable parameters for customizing their response to end-user questions. To modify these values, open the OpenAI or Google Gemini API module and modify the Inputs & Outputs settings of the pluginOpenAI or pluginGoogleGCP Plug-In components.

The GenAI Connector template includes the following required and optional parameters:

Parameter Name Plug-In Inputs Field Description

Messages*

prompt (Google)

messages (OpenAI)

Information from the FE Front-End module is mapped into the messages field to be processed by the LLM Large Language Models (LLM) are artificial intelligence (AI) programs that use deep learning algorithms to analyze and generate text and other content based on large amounts of data..

*This parameter is required.

Model*

model (OpenAI)

Indicates which LLM Large Language Models (LLM) are artificial intelligence (AI) programs that use deep learning algorithms to analyze and generate text and other content based on large amounts of data. model from the provider to use. To balance cost and performance, use the model that best fits your use case.

Unqork currently supports only LLM Large Language Models (LLM) are artificial intelligence (AI) programs that use deep learning algorithms to analyze and generate text and other content based on large amounts of data. models. Generative/Multimodal Unlike LLMs (Large Language Models), Generative and Multimodal AI models can accept text or media inputs, and output text or media responses. models are not available at this time.

*This parameter is required.

Learn more about Gemini models here: https://ai.google.dev/gemini-api/docs/models/gemini

Learn more about OpenAI models here: https://platform.openai.com/docs/models/overview

Role*

  

Role is initially set to user to designate that the information sent to the LLM in the chatbot experience is sent by the end-user End-users, also known as Express Users, are the individuals accessing an application through Express View. In most cases, end-users are the customers using the product..

*This parameter is required.

Maximum Tokens

maxOutputTokens (Google)

max_tokens (OpenAI)

Set the limit on token usage per request.

Learn more about Gemini tokens here: https://ai.google.dev/gemini-api/docs/tokens

Learn more about ChatGPT tokens here: https://platform.openai.com/docs/introduction/tokens

Temperature

temperature (Google and OpenAI)

Restrict the LLM’s response types from creative to precise to ensure that your end users have the experience that you desire.

Learn more about Gemini Temperature here: https://ai.google.dev/gemini-api/docs/models/generative-models#model-parameters

Learn more about ChatGPT Temperature here: https://platform.openai.com/docs/guides/text-generation/how-should-i-set-the-temperature-parameter

Additional model parameters can be added in the Inputs table of the Inputs and & Outputs settings.

Using the GenAI Connector Template in Express View

After installing and configuring the GenAI Connector, preview it in Express View to use the LLM.

To use the GenAI Connector in Express View:

1. From the AI Application page, click ···. The application's Options menu displays.
2. Select Preview. A new browser page displays the application.
3. Depending on what model you set up, click Google Gemini or OpenAI.

You can set up one or both of the available GenAI models.

4. Enter a message in the prompt.
5. Click the send button. The GenAI model responds to the message prompt.

Resources