GPT-6.2 Astra vs. Claude Opus 5.5 on 48 physics-game decisions

Chronological Source Flow
Back

AI Fusion Summary

A comparison of GPT-6.2 Astra and Claude Opus 5.5 across 48 physics-game decisions in Pudding Escape: Super Jelly showed Astra averaging 98.4 m versus Claude's 92.6 m, though the difference was statistically insignificant. GPT-6.2 Sol scored 69.9% on SimpleBench, below GPT-6.1's 82.9%. Per Artificial Analysis, both new models answer more concisely, with Astra cutting answer length 40 percent.
Community Comments
Loading updates...
0