Kimi K3, Opus 5, Grok 4.6, Qwen 3.8 27B, GPT 5.6 SOL: Recreate Ghost of Tsushima
I put 5 AIs against each other to see which can build the prettiest experience of a samurai roaming an ancient island of Japan (all one shotted except for one model) Kimi K3 - ran for 6+ hours and
Seeing a lot of high end blender stuff so I'm bringing back arthouse with none other than TAROT PEPE.
Lovingly conjured with Fable 5.1 and Opus 5. —>
Built the same tennis game in 2 different styles 🎾
Left — Sketchy Toon, made with Opus 5 Right — Pastel Toony, made with GPT-6 Which one would you play in ur browser? 👀
Two engineering teams just spent a quarter evaluating whether to standardize on Claude Opus 5 or GPT-6 Astra for their enterprise backend.
Analysts published deep dives comparing eleven variants across benchmarks, context windows, and per-token pricing. Everyone assumed that
You have reached the end of the archive
All of Claude Opus 5