GPT-6 Astra builds a playable 3D samurai scene from one prompt, calling Hyper3D Rodin MCP for assets

JaynitMakwana · x · 2026-09-14

Developer JaynitMakwana demoed GPT-6 Astra building a playable browser 3D scene from a single prompt: the model wrote the code, designed the level, and autonomously called the Hyper3D Rodin MCP to generate the samurai character — no Blender, no asset store.

The follow-up posts outline the workflow: inside Codex, the prompt asked for an interactive 3D samurai-in-a-forest scene with camera controls and cinematic lighting. Astra handled planning, code, and iteration, while Hyper3D Rodin MCP handled 3D asset generation — the split of "model handles logic and code, MCP tool handles assets" producing a working project rather than a placeholder demo.

Related event: GPT-6 Astra Builds Playable 3D Samurai Game From One Prompt(2 posts)→

Original post →

More from coding & agent

coding & agent channel →