I tested Claude Pro ($20)

I used Opus 5 at MED settings (forgot to use max) I am absolutely amazed!!! 20 mins is what I was thinking... here is what I actually got: - 44 minutes of usage - 58% of 5h limit used (7% weekly!) - $25 API usage - 32 Million tokens!!!

shownotoverSep 1998

JEV is king at categorization!

🚨 JEV beats the TOP embedder (Qwen3-Embedding-4B) at categorization! 46% vs 97% accuracy! Not a simple win! We benchmarked +1000 TODO titles: - JEV scored 97% accuracy compared to Opus 5 reference. Qwen3 only 46%! - Embedder costed

Marcell HavlikSep 198

Claude Opus 5 vs Gemini 4 vs GPT-6 Astra vs Grok 4.6

I handed the same unfinished drawing to four different AI models and said "be creative." Rumoured Gemini 4 checkpoint is inside Arena. four completely different pandas came out. and I genuinely can't pick a favorite 👇

Poonam SoniSep 1927

Gemini 4 pro VS opus 5 🤯

Arena testers ran the same Three.js 3D prompt across both models: • Opus 5: Basic low-poly mesh • Gemini 4 Pro: Coherent spatial depth & mesh topology • Full interactive WebGL export Google is dominating 3D code.

Web3BitsSep 19