Image Generate

Image Generate is the main generative image node. It combines a model, prompt, reference attachments, output geometry, and variation count in one place. Its output can land in Image Preview nodes or continue into another image tool.

The available controls change with the selected model. A model that does not support references, quality tiers, custom dimensions, or several variants does not show those controls.

1. 1

   Add Image Generate

   Find Image Generate in the Generative group and add it to the canvas.
2. 2

   Choose a model

   Open the model menu. Models are grouped by provider and show the controls they support.
3. 3

   Write the prompt

   Describe the subject, setting, light, composition, and finish that matter to this result.
4. 4

   Set the output

   Choose the ratio or dimensions, resolution, quality, and number of variants offered by the model.
5. 5

   Press Run

   Review the credit estimate on the button, then run the node. The editor creates missing Image Preview boxes automatically.

Model

Chooses the image model and determines every capability shown below it.

Prompt

Describes the image. A connected text input makes this field read-only because the upstream node owns the text.

Attachments

Adds uploaded or wired reference images when the model supports them.

Ratio or dimensions

Sets the output shape. Pixel-sized models can also offer custom width and height.

Resolution and quality

Appears only for models that expose those choices. Some options change the credit cost.

Variants

Requests several independent results from one press, up to the selected model's limit.

Enhance

Improves a non-empty local prompt for the selected model. It is unavailable when a wire owns the prompt.

Run mode

Chooses whether a run adds results or overwrites the wired previews.

Add a reference by uploading it inside the node or wiring an Image Input, Image Preview, or another compatible image output into Attachments. References supplement the prompt and can help preserve a product, person, composition, or visual direction.

Switching models keeps references only when the next model accepts them. If its limit is lower, the node keeps only the number it supports.

Write a name inside square brackets, such as [Product name]. The node creates a text input port for that variable. Connect a Text Input or Text Generate node to supply its value at run time.

Each unique variable gets one port. This is useful when the prompt structure should remain fixed while a product name, material, location, or campaign line changes.

A normal run fills empty wired previews first. If the node requests more variants than it has empty previews, the editor creates the missing boxes and wires them automatically.

Run Overwrite replaces completed results in the wired previews. Use it when those preview holders already feed later steps and their connections need to remain intact.

• Text Input → Prompt

  Keep a reusable prompt or campaign direction outside the generator.

• Image Input → Attachments

  Use an uploaded product, subject, or style reference.

• Image Generate → Image Preview

  Choose the exact holder that receives a result.

• Image Preview → Image Generate

  Use a completed result as a reference for the next generation.

• Image Generate trigger → another generator

  Start the next paid step only after this one succeeds.