Procedural cheetah in Three.js: GPT 6.1 Sol Max vs Opus 5.5 Max
majidmanzarpour · x · 2026-09-30
The author ran a hands-on coding test building a procedural cheetah in Three.js, comparing the outputs of GPT 6.1 Sol Max (left) vs Opus 5.5 Max (right), with side-by-side results attached.
More from coding & agent
- Swargs Rust v0.3.0 ships sub-agents, handoffs, and typed tool outputs — KyeGomezB · 2026-09-30
- Fixing Qwen-Image 2.1 pixel drift: just two ComfyUI nodes, no full workflow needed — MEDFAX · 2026-09-30
- Agent engineering rule: build the slow fallback that always finishes; fast path is an optimization — victor_explore · 2026-09-30
- Hermes Agent Trick: Give Your Bot Its Own Desktop and Take Over at Login — Teknium · 2026-09-30
- Whop's team ships its mobile app with coding agents running on Revyl's cloud devices — eptwts · 2026-09-30
- GamowLabs' LabBench: AI Agents Struggle to Pick the Next Best Wet-Lab Experiment — danielmckinn0n · 2026-09-30