Claude Solves “World-Class” AI Benchmark

Do you recall the “World-Class AI Benchmark” post I created?

If you do, you might remember that I have not tried Claude, now here it is, my dad just got a Claude subscription(also have gemini and chatgpt so yeah), specifically Claude Opus 4.6:

He is correct!

Gemini 3.1 Pro is still wrong hehe.