Bindu Reddy: Sonnet 5.5 'doesn't really work' at any reasoning level
bindureddy · x · 2026-09-29
Bindu Reddy says Sonnet 5.5 falls short at every tier: standard mode is much worse than Sol 5.6, and xHigh mode underperforms on agents. He likens it to a 'Sol 6'—cheaper and faster but worse.
More from Models
- Burkov on Sonnet 5.5 High: 'crazy how fast it is' for code questions — burkov · 2026-09-29
- Why 4o feels different: thread argues native omni training, not capability, shapes model personality — RileyRalmuto · 2026-09-29
- Sonnet 5.5 effort settings make no difference in 15-task coding test: 9/15 at low, medium and high — every · 2026-09-29
- PrunaAI claims its text-to-video modes sit on DesignArena Pareto frontiers — guennemann · 2026-09-29
- Reddit user: Opus 5.5 silently falls back to Opus 5 on nearly every prompt — fishcat_catfish · 2026-09-29
- Sonnet 5.5 clones open-source editor Proof at low effort, joining elite group of just four models — every · 2026-09-29