Image Generation
Generate images using AI models like Imagen, Flux, and more.
The Image Generation node is the core of most Clickraft workflows. It runs AI models to create images from text prompts and optional reference images.
Inputs & Outputs
| Direction | Handle | Position | Type | Description |
|---|---|---|---|---|
| Input | reference-images | 30% | 🟣 Image | Optional reference images for style/composition guidance |
| Input | prompt-input | 55% | 🟢 Text | Text description of the desired image |
| Output | image-output | — | 🟣 Image | The generated image |
How It Works
- Add an Image Generation node to your canvas
- Connect a Text Prompt or Assistant to the prompt input
- Optionally connect an Image, Shopify Import, or UGC Reference node as a reference
- Select your preferred AI model from the dropdown
- Configure aspect ratio and batch count
- Click Run or press
Cmd/Ctrl + Enter
Configuration
| Setting | Description |
|---|---|
| Model | Choose from available image models — see Image Models for the full list |
| Aspect Ratio | Output dimensions: 1:1, 16:9, 9:16, 4:3, 3:4 (availability varies by model) |
| Number of Images | Generate 1–10 variations in a single run (batch generation) |
Model Selection
Models are organized into tiers:
| Tier | Characteristics |
|---|---|
| Draft | Fastest generation, lowest credit cost — great for testing prompts |
| Standard | Good balance of quality and speed for everyday use |
| Pro | Higher quality output for marketing materials |
| Premium | Best quality — use for hero images and flagship content |
Batch Generation
Set the number of images to generate up to 10 variations per run. Each image in the batch:
- Uses the same prompt and reference images
- Produces a different variation
- Costs credits independently
This is useful for creating options to choose from or generating content for multiple placements.
Reference Images
When you connect an image to the reference-images input, the AI uses it as a guide for style, composition, or content. How the reference is used depends on the model:
- Style reference — The AI matches the visual style, color palette, and mood
- Composition reference — The AI follows the layout and framing
- Subject reference — The AI incorporates elements from the reference image
Not all models support reference images equally. Check Image Models for model-specific capabilities.
Credits
Each generated image costs credits based on the selected model. Credit cost is deducted before generation starts. If generation fails, credits are automatically refunded.
See Credit System for details on how credits work.
Tips
- Start with a Draft tier model to iterate on your prompt quickly, then switch to a higher tier for the final output
- Use specific, detailed prompts — the more information you provide, the better the results
- Reference images significantly improve consistency when generating product photography
- Generate in batches to quickly compare different AI interpretations of the same prompt