Google Vertex AI icon

Google Vertex AI

Use Google Vertex AI (Gemini) for text, media analysis, and image generation/editing

Image → Generate an Image

AI-generated

Summary

Generates an image from a text prompt using a Gemini image ("Nano Banana") model on Google Vertex AI and returns it as binary data on the output item.

Inputs

  • Image Model (required) — Gemini image model to use: gemini-2.5-flash-image (default), gemini-3-pro-image, gemini-3-pro-image-preview, or gemini-3.1-flash-image.
  • Prompt (required) — Text description of the image to generate.
  • Options > Aspect Ratio — Output aspect ratio such as 1:1, 16:9, 9:16, 4:3 or 3:4. Left blank, the model default applies.
  • Options > Image Size — Output resolution: 1K, 2K or 4K. The node's description says 2K/4K are supported by the Gemini 3 image models. The default of 1K is sent if the option is not set.
  • Options > Put Output In Field — Name of the binary field that holds the generated image. Defaults to "edited". Additional images get a numeric suffix (edited1, edited2, and so on).

Output shape

One output item per input item. The generated image(s) are in the binary property (default "edited"). The json holds { text } if the model returned any accompanying text, otherwise it is empty.

If the model returns no image (for example, the request was blocked), the item has only json with text and, if present, blockReason, and no binary data. Duplicate images are dropped. Files are named like edited.png using the returned MIME type. The request uses responseModalities TEXT and IMAGE. Uses the Vertex AI service-account credential, whose project and region determine where the model is called; a model may not be available in every region, and the "global" region is suggested in the error hint. With Continue On Fail enabled, errors are returned as json.error.

Examples

Example 1: Create a 16:9 banner image from a text prompt

Image Model: gemini-2.5-flash-image; Prompt: "A minimalist banner of a mountain at sunrise"; Options > Aspect Ratio: 16:9

Example 2: Produce a high-resolution image and store it under a custom binary field for a later upload node

Image Model: gemini-3-pro-image; Image Size: 4K; Put Output In Field: cover

Links

Discussion