Claude's 'taste' beats GPT-6.1 Sol, sparking calls for a taste benchmark

AiBreakfast · x · 2026-10-02

Replying to a user who said GPT-6.1 Sol is efficient but Opus 5.5 and Sonnet 5.5 feel better, AiBreakfast quipped: "We need a benchmark for taste / vibes." The original poster noted Anthropic has brought back Claude's great "taste" and the current models are fast, efficient and simply very good.

Related event: KOLs Prefer Claude Opus 5.5's "Taste" Over Efficient GPT-6.1 Sol(7 posts)→

Original post →

More from Models

Models channel →