Skip to main content
All CollectionsIntegrations and EnrichmentsOpenAI/GPT
ChatGPT: Analyze image with OpenAI/GPT
ChatGPT: Analyze image with OpenAI/GPT

Ask ChatGPT a question relating to images using the GPT Vision models.

Updated over 3 months ago

Introduction

This integration enables users to leverage OpenAI's GPT Vision models within Clay's platform to analyze images. By utilizing the ChatGPT: Analyze image action, users can ask questions about visual content, extracting insights and information from uploaded images. This functionality combines Clay's data capabilities with OpenAI's advanced image understanding technology, enhancing the ability to process and interpret visual data within workflows.

Input

Name

Is Optional

Description

Type

Prompt

The prompt message to be sent to ChatGPT, in addition to the images.

longtext

Webpages to Screenshot

true

URLs of the webpages to be screenshotted and then sent to ChatGPT, separated by commas.

text

Should screenshot from a mobile device?

true

By default, the screenshot will capture as if taken from a desktop. Toggle this on if the webpage should be screenshotted with a mobile device aspect ratio.

boolean

Should screenshot full page?

true

By default, the screenshot will only capture the portion of the page that is visible in the browser's viewport. Toggle this on if the webpage should be screenshotted in full page mode -- since the AI model will resize the image with increased dimensions, this MAY REDUCE QUALITY, especially for long scrollable pages.

boolean

Raw Image URLs

true

URLs of images to be sent to ChatGPT, separated by commas. These must be pure image URLs, not webpages.

text

Creativity Level

true

Choose the creativity level that GPT can have with the given prompt, higher temperatures being higher creativity / variance. You can [read more about it here.](https://platform.openai.com/docs/api-reference/chat/create#chat-create-temperature)

text

Output

Name

Type

Response

text

Did this answer your question?