GPT-6 Astra vs Claude Opus 5 is getting ridiculous
Astra : → 99.9% on ARC-AGI-3 → 98% on FrontierMath Tier 4 → $10 input / $50 output per 1M tokens Claude Opus 5 : → SOTA on Frontier-Bench → SOTA on GDPval-AA → ~1.5x the next-best model on AutomationBench at the same
This is how I build education using Opus 5 at Claude Code (Step by Step) ⬇️
Idea - User Flow - Wireframes - First Draft - Iterations - Final Design I want more details here ⬇️
Claude Opus 5. Social intelligence with Nooticr.
It took Claude a few minutes (3 minutes) to figure out that Stalkr was created by marclou The prompt: Stalkr is one of our competitor i want to see what they are doing on X (Social) Try it yourself:
Say what? 6 months ago these ski videos were called "montage".
Today, no one even stops to look. Alex Ermolov's snowboarder test is actually based on a simple question: Can an artificial intelligence agent create a small 3D snowboarding game from a single text? It's not just the view.
You have reached the end of the archive
All of Claude Opus 5