Nano Banana Editor
This project allows users to generate new images from text descriptions or edit existing ones using AI within Drupal's media system. It supports combining multiple images, offers different AI models for speed or quality, and provides control over aspect ratio and resolution.
Features
- Edit Existing Images: Transform media images with AI-powered editing
- Generate New Images: Create images from text descriptions or image references
- Multi-Image Composition: Combine up to 14 images (model-dependent)
- Model Selection: Choose between Gemini 2.5 Flash (fast) and Gemini 3 Pro (high quality)
- Aspect Ratio Control: 1:1, 3:4, 4:3, 9:16, 16:9, etc.
- High-Resolution Output: Up to 4K resolution with Gemini 3 Pro
- Style Presets: Create reusable style prompts for consistent image generation
- System Instructions: Set global prompt behavior for all generations
- Media Integration: Edit/Generate buttons directly in Media forms
- Thumbnail Previews: Visual previews of uploaded images
- Intuitive UI: Sidebar layout with configuration options and live preview
Intro Video from Drupal Developer Days Greece Keynote
Post-Installation
Ensure ai_provider_nanobanana is installed and configured.
Configuration
Navigate to Configuration → Media → NanoBanana Settings (/admin/config/media/nanobanana)
System Instructions
Global instructions prepended to all prompts. Use this to set consistent behavior, tone, or quality standards.
Example:
Generate high-quality, professional images suitable for editorial use. Maintain photorealistic style unless otherwise specified.
Image Styles
Create reusable style presets that users can select when generating images.
To add a style:
- Click Add style button
- Enter a Style Name (e.g., "Watercolor", "Photorealistic", "Comic Book")
- Enter Style Prompt (e.g., "in watercolor painting style", "photorealistic with dramatic lighting")
- Click Add style
Usage
Generating New Images
- Navigate to Content → Media → Add Media → Image (
/media/add/image) - Click Generate with NanoBanana button
- The generator interface opens with:
- Prompt: Describe the image you want to create
- Model: Select Gemini 2.5 Flash or Gemini 3 Pro
- Images (optional): Upload reference images (up to 3 for Flash, 14 for Pro)
- Image Size (Gemini 3 Pro only): Choose resolution
- Aspect Ratio: Select output dimensions
- Style: Apply a preset style (optional)
- Click Generate to preview the result
- Click Save to create a new media entity with the generated image
Example Prompts:
- "A serene mountain landscape at sunset, photorealistic"
- "Modern minimalist office interior, bright natural lighting"
- "Portrait of a golden retriever puppy, shallow depth of field"
- "Abstract geometric pattern in blue and orange tones"
Multi-Image Composition
Combine multiple images to create composite scenes, style transfers, or complex compositions.
To use:
- In either Edit or Generate form
- Select model (Flash: up to 3 total images, Pro: up to 14 total images)
- Upload additional images in the Additional Images field
- Enter a prompt describing how to combine them
Example Use Cases:
- Style transfer: Upload a content image and a style reference
- Object composition: Combine objects from different images
- Scene blending: Merge multiple scenes into one cohesive image
- Character combinations: Compose characters from different sources
Example Prompt:
Combine these images into a single scene with the character from image 1 in the environment from image 2, maintaining consistent lighting
Editing Existing Images
- Navigate to any media entity edit form (e.g.,
/media/1/edit) - Click Edit with NanoBanana button
- The edit interface opens with:
- Current Image: Preview of the existing image
- Prompt: Describe the changes you want to make
- Model: Select Gemini 2.5 Flash or Gemini 3 Pro
- Additional Images (optional): Upload up to 13 additional images to compose with the original
- Image Size (Gemini 3 Pro only): Choose resolution (Default, 1K, 2K, 4K)
- Aspect Ratio: Select output dimensions
- Style: Apply a preset style (optional)
- Click Generate to preview the result
- Click Save to replace the media entity's image, or Cancel to discard
Example Prompts:
- "Convert this to a watercolor painting"
- "Remove the background and make it transparent"
- "Change the sky to sunset"
- "Add dramatic lighting from the left"
Model Comparison
Gemini 2.5 Flash Image
- Speed: Fast, low-latency generation
- Multi-Image: Up to 3 total images (1 main + 2 additional in Edit mode)
- Resolution: Standard
- Best For: Quick iterations, real-time editing, prototyping
Gemini 3 Pro Image (Preview)
- Speed: Slower, higher quality
- Multi-Image: Up to 14 total images (1 main + 13 additional in Edit mode)
- Resolution: Default, 1K, 2K, 4K
- Best For: Final production, high-quality output, complex compositions
Additional Requirements
- Drupal 10.3+ or Drupal 11+
- AI module
- ai_provider_nanobanana module
- Media module (core)
- Google Gemini API key configured in ai_provider_nanobanana