---

name: gemini-image-edit description: Edit existing images with text prompts using fal.ai Gemini 3 Pro. Use when the user wants to modify, edit, transform, or change an existing image based on a text description. Supports multiple input images for context. allowed-tools: Bash, Read, Write

Gemini Image Editing

Edit and transform existing images using text prompts with Google's Gemini 3 Pro model via fal.ai.

Prerequisites

• FAL_KEY environment variable must be set (typically in ~/.zshrc)

API Endpoint

POST https://fal.run/fal-ai/gemini-3-pro-image-preview/edit

Parameters

Required

• prompt (string): The editing instruction describing what changes to make
• image_urls (array of strings): URLs of the images to edit

Optional

Usage

cURL

Copycurl --request POST \
  --url https://fal.run/fal-ai/gemini-3-pro-image-preview/edit \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "prompt": "Add snow to this mountain scene and make it winter",
    "image_urls": ["https://example.com/mountain.jpg"],
    "num_images": 1,
    "output_format": "png"
  }'

Python

Copyimport fal_client

result = fal_client.subscribe(
    "fal-ai/gemini-3-pro-image-preview/edit",
    arguments={
        "prompt": "Add snow to this mountain scene and make it winter",
        "image_urls": ["https://example.com/mountain.jpg"],
        "num_images": 1
    }
)

# Access the edited image URL
edited_url = result["images"][0]["url"]
print(f"Edited image: {edited_url}")

JavaScript

Copyimport { fal } from "@fal-ai/client";

const result = await fal.subscribe("fal-ai/gemini-3-pro-image-preview/edit", {
  input: {
    prompt: "Add snow to this mountain scene and make it winter",
    image_urls: ["https://example.com/mountain.jpg"],
    num_images: 1
  }
});

console.log("Edited image:", result.images[0].url);

Response Format

Copy{
  "images": [
    {
      "file_name": "edited_image.png",
      "content_type": "image/png",
      "url": "https://storage.googleapis.com/..."
    }
  ],
  "description": "A description of the edited image"
}

Examples

1. Style transformation:

   • Prompt: "Convert this photo to a watercolor painting style"

2. Object addition:

   • Prompt: "Add a rainbow in the sky"

3. Scene modification:

   • Prompt: "Change the time of day to sunset with golden hour lighting"

4. Multiple reference images:

   • Prompt: "Combine elements from these images into a cohesive scene"
   • Provide multiple URLs in image_urls array

Tips

• Be specific about what changes you want
• Use descriptive language for style changes
• Multiple images can be provided for context or combining elements
• The auto aspect ratio preserves the original image proportions

Error Handling