Post-Cutoff.com
  1. Home
  2. Videos
  3. Third iteration of an Opus 5.5 web-swinging…

Third iteration of an Opus 5.5 web-swinging game: Blender, image generation and three.js in the browser (X video)

Shikhar (@xikhar) · 2026-09-27 · ai-made · 1,796,321 views

▶ Watch on X

Made by AI

Model: Claude Opus 5.5, image-generation model (not named) · Series: Agent-built game or 3D world (the video shows the result)

Evidence: X post (2026-09-27): 'Third iteration with Opus 5.5 medium. / / I am in awe. It turned blender, image-gen, and three.js into this beauty, which runs on your browser. / / An understatement to say that Anthropic cooked. You are looking at the future of game dev.'

Human role: Iterated three times with Opus 5.5 (medium effort); the model combined Blender, image generation and three.js. Prompts not published.

Pipeline: Opus 5.5 → Blender assets + generated textures + three.js game → 2:21 gameplay video

What's in the video

Description written by Gemini, which watched and listened to the whole video.

Summary
This video showcases gameplay footage of a third iteration of a browser-based 3D web-swinging game created with assistance from Claude Opus 5.5, utilizing Blender, AI image generation, and Three.js. Shared by Shikhar (@xikhar), the demo highlights traversal physics, dynamic camera work, and urban exploration in both classic red-and-blue and black symbiote suits.

What is shown

Claims & numbers

Notable quotes

Assessment
This is a real technical gameplay demonstration showcasing an impressive hobbyist/indie 3D browser game prototype built using Three.js and AI tooling. The captured gameplay is real-time footage displaying working physics, animation blending, collision detection, and procedural city traffic.

Lyrics & themes

Lore & references

Visual style & craft
The video displays real-time 3D WebGL graphics rendered in a web browser using Three.js. The environments combine modular city blocks modeled in Blender with image-generated storefronts and billboard textures. Character models feature full skeletal rigging, directional web attachments to building colliders, dynamic camera shake and FOV widening during terminal velocity dives, and shadow mapping across street-level traffic.

Described by gemini-3.8-flash on 2026-09-30 from the video's audio and frames.

Related events