Plugin Mode

Generate Text from Image with OpenAI

Summary

Click the button to send an image file and text (a multi-line text field) to OpenAI, and set the result into a multi-line text field.

Smartphone support

This plugin does not support smartphones. Use it on a computer.

How to use

Choose the attachment field for the image and the fields for the prompt and response. Configure the connection and button label, then attach an image, enter a prompt, and click the button.

When it runs

Depending on the settings, this customization runs at the following times. Any configured conditions are also evaluated.

  • When you open a record details page
  • When you click a button added by this plugin

Click the button added by this customization to run the operation.

Parameters

Enter or select a value for each setting unless it is marked optional.

Prompt Input Field

Choose the field where users enter the prompt to send to the AI.

The following field types can be selected.

  • Text area

Fields inside tables cannot be selected.

Label

Enter a button label that makes its purpose clear to users.

Initial value: Button

Image Input Field

Choose the attachment field for images to be read by the AI.

The following field types can be selected.

  • Attachment

Fields inside tables cannot be selected.

OpenAI API key

Enter the API key used to connect to OpenAI.

Language Model

Specify the language model to use for the operation.

Available options:

  • GPT-5.6 Sol
  • GPT-5.6 Terra
  • GPT-5.6 Luna
  • GPT-5.5
  • GPT-5.2
  • GPT-5
  • GPT-5 mini
  • GPT-5 nano
  • GPT-4o
  • GPT-4o mini
  • GPT-4 Turbo

Initial value: GPT-5.6 Sol

Destination Field

Please select the field where you want to set the response from OpenAI.

The following field types can be selected.

  • Text area

Fields inside tables cannot be selected.

Restrictions

Operation is not guaranteed in apps where the field code of the Record number field has been changed.

The following actions and conditions are useful when building similar behavior in customization mode.

Actions

Conditions