Karpathy Tests Claude Opus: Writes 5,500 Lines of Code in 2 Hours to 3D Render LOTR
andrewchen · x · 2026-08-02
Andrej Karpathy shared an exploration into the evolving ways we evaluate LLMs, noting that we are moving past simple tests like "generate an SVG of a pelican on a bicycle."
To test long-horizon task capabilities, he provided Claude 3 Opus with the opening paragraph of The Lord of the Rings, a 1 million token budget ($10), and asked it to render the scene using Three.js. Opus ran for about 2 hours and generated 5,500 lines of code that procedurally rendered the story. Karpathy found it mindboggling that the LLM could place and orchestrate polygon assets in (x,y,z) coordinates and write the animation code to make it actually work. He also highlighted a new paradigm: highly custom, tedious code that no human would ever have the patience to write manually can now be delegated to LLMs for practically "free".
Related event: Karpathy's LOTR 3D test pushes long-horizon model evaluation(7 posts)→
More from coding & agent
- ERP•AI launches headless SaaS with $0 seats that bills machines: $1 buys 4,000 record writes — rohanpaul_ai · 2026-10-16
- FDE-Gym benchmark shows huge gap on real AI delivery tasks: top model scores 0.563 — DeepLethe · 2026-10-16
- Vibe-coded dashboard leaked franchise financials before anyone logged in — shauntrennery · 2026-10-16
- Fintech HR built a sourcing agent on 60,000 past applicants — 30 qualified candidates in a month — G0NZ0BOY · 2026-10-16
- Orchestrating a 100-image portrait of humanity for aliens with ChatGPT, Argon and p5.brush — dx8xb · 2026-10-16
- 3x DGX Spark cluster runs GLM-5.3-Flash locally behind a proxy, 1.75M tokens/day at 119W — Teknium · 2026-10-16