Analyze Image
The Analyze Image block sends an image to a vision model along with a text prompt, and returns a text-based analysis of that image.
Analyze images using a vision model and prompt
The Analyze Image block sends an image to a vision model along with a text prompt, and returns a text-based analysis of that image. It takes two primary inputs: a prompt that describes what to look for or extract, and a URL pointing to the image to be analyzed. The result is stored as a string in an output field called analysis, and can optionally be saved to a named variable for use later in the workflow.
The block uses the workflow's default vision model unless a model override is specified via the optional visionModelOverride setting. This makes it straightforward to swap in a different vision model for specific steps without reconfiguring the entire workflow.
This block fits into workflows that need to extract information from images, such as reading text in a photo, describing visual content, or checking an image against specific criteria. It can be connected to upstream blocks that supply image URLs and downstream blocks that act on the resulting analysis text.
What you can build
Real-world workflows powered by the Analyze Image block.
Product Image Review
Analyze product photos submitted by sellers to verify they meet listing guidelines before publishing to a catalog.
Document Data Extraction
Extract structured information from scanned documents or forms by prompting the vision model to identify specific fields.
Accessibility Alt Text Generation
Automatically generate descriptive alt text for images uploaded to a content management workflow.
Visual Quality Control
Inspect images from a manufacturing or fulfillment process to detect visible defects or anomalies based on a defined prompt.
Social Media Moderation
Review user-submitted images against content policy criteria and flag those that may require human review.
Receipt and Invoice Parsing
Analyze photos of receipts or invoices to extract line items, totals, and vendor details for expense tracking workflows.
Ready to add Analyze Image to your workflow?
Get Started FreeCommon questions about Analyze Image
What are the required parameters for this block?
The block requires two parameters: a prompt (a string describing how to analyze the image) and an imageUrl (a string containing the URL of the image to analyze).
What does the block return?
The block returns a single output field called analysis, which is a text string containing the vision model's response based on the provided prompt and image.
How is the analysis result stored for use in later steps?
You can specify an optional destinationVar parameter to save the analysis text to a named variable, which can then be referenced by downstream blocks in the workflow.
Which vision model does the block use?
By default, the block uses the vision model configured at the workflow level. You can override this on a per-block basis using the optional visionModelOverride setting.
What kinds of workflows is this block commonly used in?
This block is used in workflows that need to interpret or extract information from images, such as document processing, content moderation, product review, and automated reporting pipelines that receive image URLs as inputs.
Related capabilities
Add Analyze Image to your workflow
Build powerful AI workflows with drag-and-drop blocks. No coding required.