Hey everyone — I’ve been building Quadify tools over the past year, and today I’m launching Quadify Pro, the mid-tier that sits between the free Base tool and the ML-powered Ultra.
Here’s an honest breakdown of which tool is for you:
QUADIFY BASE ($10)
One-button tri-to-quad conversion. Good for simple meshes and learning the workflow. No pipeline tools, no Agent Mode.
QUADIFY PREMIUM ($25) — production asset optimization for environment artists. Density Target (tris/m² rather than arbitrary ratios), region-aware protection, LOD generation, batch across prop libraries. The tool for world-building pipelines.
QUADIFY PRO ($50) ← this one
Full production pipeline. Multi-pass quadification with selection-based targeting so only the faces you pick get processed. Includes:
• LOD generation with face count preview per level
• UV unwrap + UDIM packing
• Batch processing across multiple objects
• Texel density normalization
• Collision mesh generation
• QA heatmaps and HTML reports
• AI Agent Mode (19 operations)
QUADIFY ULTRA ($150)
ML-powered one-click retopology with algorithm routing, Reference-Driven Retopology, Loop Flow Intelligence, and Animatability Score. Better for complex organic meshes where topology flow matters more than pipeline control.
WHEN IS PRO THE RIGHT CHOICE?
Game asset pipeline: You need quads + LODs + collision + UV in one pass without switching tools. Pro handles this as a connected pipeline.
Photogrammetry/scan data: Dense, messy imports where you want precise control over which faces get processed. Selection-based targeting is the core feature.
Natural language automation: Type “convert to quads, reduce to 15k faces, generate 3 LODs, UV unwrap, export FBX” — Agent Mode plans and runs all 5 steps with a safety preview before anything executes.
When to use Ultra instead: Complex organic characters or faces where edge loop flow matters more than pipeline automation.
ABOUT AGENT MODE
Pro’s Agent Mode reads your actual mesh data (face count, quad %, non-manifold edges, shape keys, armature parents) before planning anything — it doesn’t guess. You get a full step-by-step plan with confidence scores and destructive flags before a single vertex moves.
Runs on your own Claude (Anthropic) or GPT-4o (OpenAI) API key. ~$0.001–0.003 per run. Only your instruction text and anonymous mesh stats are ever transmitted.
Links:
ArtStation: https://www.artstation.com/a/54349362
Superhive: https://superhivemarket.com/products/quadify
Happy to answer questions about which tier fits your workflow. If you’ve used Base and want to see the full pipeline, DM me.