Max reasoning effort nearly doubles cost for minimal Intelligence Index gains
randal_olson · x · 2026-10-07
Using Artificial Analysis's Oct 6 snapshot, Randal Olson shows max reasoning effort nearly doubles cost per benchmark task for minimal Intelligence Index gains: Claude Opus 5.5 at xhigh ($3.46/task, score 56) roughly matches Sonnet 5.5 at max ($7.67/task, 56) for less than half the cost. Full ladder: Opus 5.5 max $5.98 (58), Sonnet 5.5 max $7.67 (56), Opus xhigh $3.46 (56), Opus high $1.82 (54). Takeaway: compare models before switching to max. Charts made with his open source evident-charts skill.
Related event: Max Reasoning Effort Nearly Doubles Cost for Minimal Gains(2 posts)→
More from Models
- Dev warns OpenRouter share, cache hit rate, latency stats are easily gamed for marketing — charles_irl · 2026-10-07
- OpenAI's unreleased model reportedly proves quasi-Riemann hypothesis with Lean proof — ChrisGPT · 2026-10-07
- Fed Claude Opus my blurry handheld Saturn shots, it fused them into one best image — adonis_singh · 2026-10-07
- Two Labs, One Race: Anthropic vs OpenAI Frontier Model Release Timeline, 2023–2026 — Medical-Sky7620 · 2026-10-07
- Claude was given robot skin — Opus was curious but anxious about hooking up — repligate · 2026-10-07
- Gemini 2.5 Pro retiring October 20, 2026, users say goodbye — hargup13 · 2026-10-07