Welcome to VariantLab
VariantLab is an AI-powered procedural art and asset generator that transforms your AI-generated images into scalable asset collections. Create thousands of unique variations from a single design.
The Power of Trait-Based Art
VariantLab specializes in creating trait art - artwork composed of interchangeable visual elements (traits) that can be mixed and matched to create unique combinations. This is the same technique used to create popular NFT collections, game character systems, and avatar generators.
What Are Traits and Layers?
- Traits are distinct visual elements of your artwork (eyes, hair, accessories, backgrounds, etc.)
- Layers are transparent PNG images of each trait variation that stack together
- Combinations are unique compositions created by selecting one layer per trait
For example, a character with 5 eye variations, 4 hair styles, and 3 backgrounds can produce 60 unique combinations (5 × 4 × 3) - all from just 12 images.
Exponential Output, Minimal AI Spend
Traditional AI art generation requires generating every single image individually. For a 1,000 piece collection, that's 1,000 AI generations.
VariantLab flips this equation:
| Approach | Total Images | AI Generations | Cost |
|---|---|---|---|
| Traditional | 1,000 | 1,000 | ~$15-40 |
| VariantLab | 1,000 | ~30 | ~$5-10 |
With just 5 traits and 5 variations each, you get:
- 25 AI-generated variations + 5 originals = 30 total generations
- Output: 5 × 5 × 5 × 5 × 5 = 3,125 unique combinations
That's 100x more output from your AI budget.
The math is simple: instead of generating each final image, you generate a small set of layer variations and let VariantLab combine them procedurally. More traits and more variations per trait means exponentially more unique outputs.
What Can You Create?
- NFT Collections - Generate unique generative art with trait-based rarity
- Game Assets - Create character variations, items, and sprites
- Avatar Systems - Build customizable profile picture generators
- Marketing Assets - Produce themed variations for campaigns
- App Development - Create commercial-ready assets for your projects
How It Works
VariantLab uses a 5-stage pipeline to transform a single image into thousands of unique combinations:
- Setup - Write a prompt and generate base images using AI
- Detection - Identify traits (eyes, accessories, backgrounds) and create masks
- Variations - Generate alternative versions of each trait, then extract them as transparent layers
- Combine - Mix and match layers to create unique compositions
- Export - Download your collection in PNG, SVG, or PSD format
Quick Links
- Quick Start Guide - Get started in 5 minutes
- Your First Project - Complete tutorial
- Understanding Mana - Credits system explained
- Pipeline Overview - How the 5-stage workflow works
Key Features
| Feature | Description |
|---|---|
| AI Image Generation | Generate images with the AI model of your choice |
| Auto Trait Detection | AI-powered detection of image regions with mask editing |
| Transparent Layer Extraction | Automatically extract PNG layers with perfect alpha channels |
| Variation Generation | Create alternative versions of any trait |
| Procedural Combinations | Generate thousands of unique combinations automatically |
| Multiple Export Formats | PNG, SVG, or PSD with layers preserved |
Getting Help
- Browse this documentation using the sidebar
- Contact us for support
- Check Pricing for plan details