Researcher's early take: opus 5.5 best since 4.5, but 5.6-sol still the best all-around model this year
DimitrisPapail · x · 2026-09-23
Researcher Dimitris Papail shares early hands-on impressions: opus 5.5 looks very good — likely the best Opus since 4.5 — but 5.6-sol remains his pick as the best all-around model of the year. He finds sol writes better and shows no measurable regression elsewhere, including math. He may update his assessment soon. Informal impressions, not a formal benchmark.
Related event: Hands-On: Opus 5.5 Best Since 4.5, But Sol 5.6 Still Tops Overall(2 posts)→
More from Models
- Voice Agents Can't Tell Who's Talking: Voice Isolation Cuts WER from 23.3% to 6.2% — thetripathi58 · 2026-09-24
- Grok bot hits 418,000 weekly users, up 24% WoW, per FT report touted by Musk — elonmusk · 2026-09-24
- Allie Miller: Claude Opus 5.5 is the first to pass her code-meets-poetry test — alliekmiller · 2026-09-24
- Dev critique: step-like gains on a driving benchmark are misleading — ben_j_todd · 2026-09-24
- Early Opus 5.5 hands-on: noticeably better writing, tone, and iOS animation polish — signulll · 2026-09-24
- Tip: rewrite your custom instructions for smarter models like Sol 6 — thederbiedone · 2026-09-24