Ini benar-benar tambang emas

Anthropic just shared one of the best tricks for Claude Code. The idea: run Sonnet, Haiku 5.5 and Opus 5.5 at once, each one doing only the job it's best at. Better results and way fewer tokens, because every task lands on the right model.

xsmryzeNov 955

Fabel 5.2 diluncurkan hari ini.

My feed says "best model ever", so I tested it against Opus 5.5. 17 coding tasks. 68 scored runs. 5 tasks were real commits from my production app. Same quality, about half the API cost. Numbers and videos below 👇

Sebastian MatysikNov 92

Ini bukan Fabel 5.2! Semua yang Anda lihat (selain truk cyber Tesla) diambil dengan Opus 5.5 sebelum di-nerf.

My prompt was very simple: Remake GTA2 game with modern graphics, everything made with code. It worked for 5H and then I was blown away by the result.

Rising Sun InteractiveNov 917

Ini benar-benar gila. Insinyur antropik Andrej Karpathy

"The biggest mistake in AI right now is forcing agents to work before you master the harness around the model We made that mistake at OpenAI in 2016. It cost us 5 years" In just 6 minutes, he explains why the agent is

kobaNov 917