Skip to main content

Prompt images (Vision)

Use the Prompt images (Vision) tool to analyze URLs in a column using AI vision capabilities. It is the only tool that accepts images as input.

For example, you can generate listing descriptions for a set of real-estate property plans.

Prerequisites
  1. Click GPT for Excel Word in the Home tab. If you don't see it, go to Home > Add-ins > My add-ins > GPT for Excel Word.

  2. Click Prompt images (Vision) from the Bulk tools tab.



  3. Configure the Prompt images (Vision) tool to match the text in your sheet:
    1. Enter the Prompt to apply to your column. You can ask a question about the image, or ask for a description of it.
    2. Select the column containing the URLs of the images you want to analyze. Supported formats are PNG, JPEG, WEBP, and GIF.
    3. Select the column to Put results in.

    4. The image shows steps to select a column with image data, enter a prompt, and choose a results column.

  4. Note: Regardless of the model you choose, this function always uses gpt-4o. If you have access to gpt-4o with your own API key, select it in the model switcher to reduce costs. Learn more about pricing.

  5. Select if you want to run a specific number of rows or All rows. Then, click Run rows.

The listing descriptions of the property plans are generated in the selected column.