Karpathy Tests Claude 3 Opus: Writes 5,500 Lines of 3D Code in 2 Hours for $10
TAbrodi · x · 2026-08-02
Andrej Karpathy shared an experiment highlighting LLM capabilities in long-horizon coding and autonomous execution. He provided the model with the opening paragraph of The Lord of the Rings, alongside a budget of roughly 1 million tokens ($10), and asked it to generate a Three.js 3D render.
The model ran for about 2 hours and autonomously wrote 5,500 lines of code to procedurally render the story. Karpathy noted it was mind-boggling that the LLM could place and orchestrate 3D polygon assets in (x,y,z) coordinates and animate them, demonstrating capabilities far beyond traditional simple coding benchmarks.
Related event: Karpathy Tests LLM with 1M Tokens to Render 3D Lord of the Rings(6 posts)→
More from coding & agent
- PullMD: Open-Source Tool to Convert Web URLs & Files to Clean Markdown — tom_doerr · 2026-08-03
- 10,000-Hour Programmers Embrace AI: Never Writing Manual Code Again — georgemillo · 2026-08-03
- AI Agents Self-Correct Causal Analysis Errors via Verifier and Cross-Review — hugobowne · 2026-08-03
- Open-Source Legal AI: 139 Agent Skills for GDPR and NDA Workflows — tom_doerr · 2026-08-03
- Prompt-Architecture as Literature: Writing Executable Text for LLMs — lnsip9reg · 2026-08-03
- Replit Backed by Founder: Zero Technical Skills Needed to Build High-Quality Products — amasad · 2026-08-03