Gemini Vertex AI icon

Gemini Vertex AI

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

Actions5

Image → Edit

AI-generated

Summary

Edits one or more images using a text prompt via Vertex AI Gemini or Imagen models.

Inputs

  • Prompt (required) — Text description of how to edit the input image(s).
  • Input Images (required) — Collection of images to edit. Each entry requires a Binary Field Name (the binary field containing the image) and a MIME Type (PNG, JPEG, WEBP, or GIF).
  • Authentication (required) — Authentication method; only Service Account is supported.
  • GCP Project ID (required) — Your Google Cloud Project ID.
  • Region — Google Cloud region for the Vertex AI endpoint (e.g., us-central1, europe-west1).
  • Model (required) — Gemini or Imagen model used for editing. Options include Gemini 2.5 Flash Image, Gemini 3.1 Flash Image Preview, Gemini 3 Pro Image Preview, Imagen 4 variants, or a custom model ID.
  • Custom Model Alias — When Model is set to Custom Alias, provide the custom model identifier.
  • Output Binary Field — Name of the binary field to store the edited image in (default: image).
  • Output Filename — Filename for the edited image (default: generated_image.png).

Output shape

json

Returns an array with one item per input image. Each item contains a binary field (named by Output Binary Field) holding the edited image and a JSON object with success status, operation, location, model, prompt, and filename.

Examples

Example 1: Edit a product photo to add a sunglasses overlay.

Set Prompt to 'Add sunglasses on the person's face', provide the input image in binary field 'data', keep default Output Binary Field 'image', and use the default Gemini 2.5 Flash Image model.

Links

Discussion