A rendering research project by Grafikui. Built to find the limits of real-time WebGL in the browser.
MOD-15 is a self-initiated R&D project by Grafikui — a design engineering practice based in London. No client brief. No deliverable. Just a question: how far can you push real-time 3D in a Next.js app without reaching for a video file?
Two experiments, one architecture. Every model is a live GLB on the GPU. Every animation is driven by scroll position. No pre-renders, no sprite sheets, no theatre.
React Three Fiber handles the render loop and scene graph. Custom GLSL shaders are injected via onBeforeCompile at material compile time. Scroll sync runs inside useFrame — no intermediary, no state updates, just window.scrollY mapped directly to uniforms and camera vectors each tick.
MOD-15 is one of several open experiments running at Grafikui — a design engineering practice building the interface layer for AI-native products.
grafikui.com→