- Model
- p-TON — the transformer model behind Image & Video.
- What it is
- A transformer-based on-model imaging model. It takes your own garment and renders it on a consistent, brand-safe model as a publish-ready photograph or a short film — your actual product on a person, at catalog quality, not a generic AI picture.
- What it does
- Turns a flat garment shot or a prompt into on-model photoshoots and 15-second video — publish-ready imagery at catalog scale, in one consistent house look.
- Inputs
- Your garment as a product or flat photo, or a text prompt — with a chosen p-TON model and setting, so every frame belongs to the same shoot.
- Outputs
- On-model photographs and 15-second videos — hero, detail and motion frames — ready to publish straight to a catalogue, storefront or campaign.
- Capabilities
- On-model photography
- 15-second video
- Brand-consistent models
- Detail & motion
- Catalog scale
- Approach
- A garment-first pipeline: your product is preserved and composed onto a consistent model and scene, so you can steer the model, pose or setting while the garment stays true to your original.
- Consistency
- The same p-TON model returns the same person and look across an entire set, so a catalogue reads as one coherent shoot rather than a pile of unrelated AI images.
- Access surfaces
- PatternAI web app
- PatternAI API
- Review
- Every render is yours to review before it goes out — the final call on what publishes to your catalogue or campaign stays with you.
