CozyClay Lets You Direct AI Video Shots in 3D Without Writing Prompts

CozyClay turns your browser into a previs studio where you block shots, pose characters, and hand the same camera moves to an AI video model.

AlphaSignalAlphaSignal
·
·
CozyClay Lets You Direct AI Video Shots in 3D Without Writing PromptsPRO
AuthorBy AlphaSignal Newsroom
Read2 min
TopicOpen Source · Video
SubtopicVideo Generation
  • CozyClay is an open source browser-based previs studio: npx cozyclay and you're in.
  • Built with Three.js and React Three Fiber, licensed AGPL-3.0, requires Node 22+.
  • Block scenes, pose characters, author camera moves, then hand the shot to an AI video model.
  • Ships an MCP server so Claude can direct the viewport in plain language.
  • Connects to NVIDIA ARDY for motion generation via an SSH-accessible GPU box.
  • Includes a token-free ARDY text encoder setup that skips the Hugging Face gate.

A new open source tool is trying to fix one of the messier problems in AI video: getting the camera to actually do what you want. CozyClay is a browser-based 3D staging studio that lets you block a scene, pose characters, and author camera moves on a timeline, then hand that exact shot to a video generation model instead of describing it in a prompt.

It runs with a single command, npx cozyclay, and opens a Three.js editor at http://127.0.0.1:5180. There's nothing to compile and no dependency tree to install, because the published npm package ships a prebuilt dist/. The project is AGPL-3.0 and built on React Three Fiber.

Previs without the Blender tax

Previs, short for previsualization, is the rough 3D pass where a shot gets blocked before anyone pays for final assets or renders. Traditionally that means Blender, Maya, or Unity plus a small mountain of setup. CozyClay collapses the workflow into a browser tab with Unity-style controls: right-drag flies, WASD walks, Alt-drag orbits the selection, and F frames it. The muscle memory carries over directly from any 3D editor.

The core loop is straightforward:

  • Stage a scene with primitives and set pieces, moved via a W/E/R gizmo, with an overhead plan view for character waypoints.

Pro article

This story is for Pro members

You've reached the end of the free preview. Upgrade to AlphaSignal Pro to read the full article - and everything else behind the paywall.

Comments

avatar