Sonnet 5.5 benchmarks near Opus 5.5 unevenly, while OpenAI keeps token-efficiency edge
cedric_chee · x · 2026-09-29
Early feedback on Sonnet 5.5 is strong: users praise its speed, clear writing, and iteration feel. Benchmarks are hard to interpret—it's close to Opus 5.5 and even wins some evals, but Opus still owns complex, open-ended work.
Two caveats from the author:
- The speed claim needs careful reading: a 30% increase in generation rate means only 23% less generation time for fixed-length output, not a 30% cut in total task duration.
- OpenAI still leads on token efficiency.
Related event: Sonnet 5.5 Nearly Matches Opus 5.5 but Sets Token Consumption Record(12 posts)→
More from Models
- Emulate-1 claims to beat AI detectors: outputs pass Pangram as human writing — alejandroll10 · 2026-09-29
- Chain-of-thought monitoring debate: an AI that knows you read its diary can deceive you with it — repligate · 2026-09-29
- Swift 1.5 + HyperQwen cuts task time 37% at 100+ tok/s on a single RTX 3090 — KingGongzilla · 2026-09-29
- Anthropic engineer: don't run Sonnet at max effort — use Opus instead — edwinarbus · 2026-09-29
- Sonnet 5.5 vs Sonnet 5: bouncing-ball physics tests from the same prompt — claudeai · 2026-09-29
- Chart: Cheaper Sol or Opus Matches Every Sonnet 5.5 Effort Level — OnAGoat · 2026-09-29