三天实测后选 Opus 5.5：GPT-6 Sol 额度好但细节失误多

Kappaemme @Kappaemme1926原推发布于 2026.09.26

在 X 上打开原推

I have a winner after 3 days with GPT‑6 Sol and Claude Opus 5.5: Opus. → Sol on High, no Fast Mode ($200 plan): 86% of my weekly limit remains. I’ve used it more heavily, and the limits are excellent. But I’ve had to correct too many missed details. → Opus 5.5 on Mid ($20 plan): 13% of my weekly limit used. The quality for that price is incredible. Opus has given me better results on almost every task I’ve tried. These aren’t benchmarks, just my experience working with both. It’s not a question of who is better, I still use them together, but if I had to choose one today, I’d pick Opus. Anthropic really delivered. Which one would you choose?