What is the Analyze Image block?

The Analyze Image block utilizes AI vision models to analyze images based on the instructions you provide.

How to configure the Analyze Image block

Once the Analyze Image block is added to your workflow, configure it using the panel on the right. Here is how you would configure the block for an application that scans an image of an invoice and captures the details.

1. Prompt: In the Prompt section, tell the AI to analyze the image based on the instructions and then add the variable name (in this case, analyze) with double curly braces.
2. Image URL: Add the variable for the uploaded image in the Image URL section. For this example, the variable is image.
3. Output: Set an output variable name. In this example, the variable name is image_output. Note that you don’t need double curly braces when inputting the variable name into this field.
4. Model Settings: Last, choose the model to analyze the image. For this example, we chose GPT-4 Turbo Vision.

Example automation workflow

Here’s what the automation workflow for this example includes:

1. User Input: The user uploads an image of a handwritten invoice.
2. User Input: The user is asked, “What would you like to know about this image?” They then provide instructions by entering: “Please provide a breakdown of what is included in this invoice.”
3. Analyze Image: The Analyze Image block takes the uploaded image and the user's instructions.
4. Display Text: The output of the Analyze Image block (the image_output variable) is displayed to the user.

‍