StyleFusion Beta

Published

StyleFusion enters beta

The core image generation pipeline went live behind Cloudflare Access. StyleFusion takes a source image, extracts its visual DNA (color palette, face geometry, texture profile, lighting, composition), and generates consistent character variations across multiple AI providers.

What shipped

  • Visual DNA extraction pipeline (palette, geometry, texture, lighting, composition analyzers)
  • Prompt compilation engine with Grimoire integration
  • Multi-provider AI routing (Google Vertex AI and additional image generation models)
  • Face geometry extraction and identity lock system
  • Identity negatives for drift prevention
  • Cloudflare R2 storage for source images and generated outputs
  • Generation queue with status tracking