Gemini Vertex AI icon

Gemini Vertex AI

Generate or edit images using Gemini 2.5 Flash Image (Nano Banana) via Vertex AI

Actions5

Image → Generate

AI-generated

Summary

Generate a new image from a text prompt using Gemini or Imagen models via Vertex AI.

Inputs

  • main (required) — Input items that trigger the image generation; each item results in one generated image.

Output shape

same length as input items, each item passed through with added data

Each output item contains the original JSON data, plus a binary field (default name 'image') holding the generated PNG image and a JSON object with metadata such as success, operation, location, model, prompt, and filename.

Examples

Example 1: Simple text-to-image generation

Set Project ID, leave Region at default, choose a model (e.g., gemini-2.5-flash-image), provide a Prompt, keep Output Binary Field as 'image'.

Example 2: Using an Imagen 4 model with aspect ratio and watermark off for seeding

Select an imagen-4.* model, set Aspect Ratio (e.g., 16:9), disable Add Watermark, provide a Seed for reproducibility, optionally adjust Image Size and Negative Prompt.

Links

Discussion