Loading pattern preview
PatternsSDK API
AI SDK Nano Banana Image Edit
AI SDK Nano Banana Image Edit
Generate and edit images using Nano Banana 2.5 Flash. Create new images from text or edit existing ones with natural language. Includes version history and image comparison.
Loading pattern preview
Problems solved
Technical challenges this implementation handles out of the box.
- Edit existing images with Gemini Flash image editing
- Apply natural-language edit instructions to an uploaded image
- Compare before and after results in the demo UI
- Keep edit workflows client-visible without a custom image backend
Use cases
Products and workflows this pattern is designed to support.
- Creative retouch assistants
- Brand asset revision tools
- Product photo cleanup workflows
- Prompt-based image editing demos
Setup
Requirements, wiring steps, and what this pattern adds to your project.
Getting started
Pick how you want to pull this pattern in. Then wire env vars and routes the same way.
- 1
Install from the preview toolbar
Copy the install command above and run it in your project — Pro patterns include a short-lived token.
- 2
Add environment variables to .env.local
GOOGLE_AI_API_KEY, AI_GATEWAY_API_KEY
- 3
Customize the agent and tool files
Adapt prompts, tools, and stop conditions for your product — Copy for AI in the toolbar helps seed that work.
- 4
Run your dev server and open the pattern route
Install dependencies if needed, then start the app and verify responses.
Environment variables
Capabilities
AI SDK APIs
generateTextgenerateImagegatewayProviders
Gemini
External services
Upstash
Files added
9 filesapp/page.tsxapp/layout.tsxapp/create-image-action.tsxapp/edit-image-action.tsxlib/rate-limit.tslib/image-utils.tshooks/use-image-version.tshooks/image-version-context.tsxREADME.md
Dependencies
8 totalnpm packages
6UpstashAISDKlucide-reactmotionNext.jsReact
Registry components
alertbutton
Critical files
app/create-image-action.tsxServer action for image creation: uses generateText with gateway and IMAGE+TEXT response modalities. Shows the gateway pattern for Gemini image generation.
app/edit-image-action.tsxServer action for image editing: processes existing image data into base64 data URL, constructs a multimodal message with image + text edit instructions, and sends to Gemini flash image model.
lib/image-utils.tsShared image processing utilities: processImageData, createDataUrl, validateFiles, extractImageData. Essential for all image generation/editing operations.
hooks/use-image-version.tsComplex state machine hook using useReducer: manages image versioning (current/previous/original), UI state, and prompts via separate reducers. Implements edit history tracking and version management.
Requirements
- Complexity
- Intermediate
- Setup time
- ~20 minutes
- Node
- 24+
- Memory
- 512MB
Upstash Redis for rate limiting
