AI Design Tool|
In Development

Frame

AI-powered graphics, from prompt to export.

Source Code

2

AI Models

98.9%

Test Coverage

PNG/JPG

Export

Overview

Frame combines AI image generation with a canvas editor for creating graphics. Generate images from text prompts using Replicate Flux or OpenAI DALL-E, add and style text overlays with drag-and-resize, use AI inpainting to selectively edit regions, and export at 1x or 2x resolution. Full undo/redo, session history, and expired URL detection.

Stack

Next.js 16React 19react-konvaReplicateOpenAITailwind 4

Features

What it does.

AI Generation

  • Text-to-image via Replicate Flux models
  • Text-to-image via OpenAI DALL-E
  • AI inpainting: select a region, describe changes
  • Session history with expired URL detection

Canvas Editor

  • Text overlays with drag, resize, and transform
  • Layer management on HTML5 Canvas via Konva.js
  • Full undo/redo for all operations
  • Export as PNG or JPG at 1x or 2x resolution
  • Copy to clipboard support

Architecture

How it's built.

1

Next.js 16 App Router

Server actions for AI API calls

2

react-konva (Konva.js)

Canvas manipulation and rendering

3

Replicate + OpenAI

Dual AI provider for generation and inpainting

4

Tailwind 4 + CVA

Styling with class variance authority

Interested in this project?

Get in touch →
← Back to all projects