Vork engine lets AI agents write shaders with a 2,500-token API
Vork reframes engine design around machine-readable APIs instead of human-centric graphics interfaces, a shift that could reshape how agents handle 3D and VFX work. Builders working on agent pipelines for rendering, simulation or generative content can compare it directly against token costs in Blender or Unreal workflows. It runs on NVIDIA GPUs via Metal and Vulkan, so teams can test it on standard cloud instances.
Developer Ravi Parikh announced Vork, a 3D and VFX engine built for AI agents rather than human users. The core API fits in roughly 2,500 tokens and lets agents write their own shaders, materials, cameras and physics. Vork is claimed to be 2x more token-efficient than Blender, Unreal Engine or Houdini, with up to 24x faster rendering. In a demo, a simulation of 1 million birds, each with unique trajectories, rendered in 97 seconds on a single NVIDIA L4 GPU, compared with 31 minutes for Blender EEVEE. The engine is written in Rust and Slang and runs on GPU via Metal and Vulkan.
绝对疯狂 Ravi Parikh @ravisparikh Dear Claude: you are being held back by human-centric graphics APIs. I've built a 3D and VFX engine where agents can write their own shaders, materials, cameras, physics. The core API fits in ~2,500 tokens. Vork is: • 2x more token-efficient vs Blender, Unreal Engine, Houdini • Up to 24x faster rendering times • Rust and Slang based engine, runs on GPU via Metal/Vulkan Below is a simulation of 1 million birds - each with unique trajectories - rendered in 97 sec on one NVIDIA L4 GPU (Blender EEVEE: 31 minutes) Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 2 🔄 0 ❤️ 1 👀 863 📊 2 ⚡
