‹ BackHN Continuity

Thread

Claude Opus 5.5 Intelligence, Performance and Price Analysis (Max)

333 points · 106 comments · theanonymousone

  1. hglaser · · focus · HN ↗
    Half the cost per task compared to Opus 5, comparing high effort to high effort. That's just really nice.

    Edit: <a href="https:&#x2F;&#x2F;artificialanalysis.ai&#x2F;models&#x2F;claude-opus-5-5?models=claude-opus-5-5-high%2Cclaude-opus-5-high#price-cost" rel="nofollow">https:&#x2F;&#x2F;artificialanalysis.ai&#x2F;models&#x2F;claude-opus-5-5?models=...

    1. makeavish · · focus · HN ↗
      Nice catch, AA only shows max effort by default and I got disappointed thinking it&#x27;s a token guzzler though: <a href="https:&#x2F;&#x2F;artificialanalysis.ai&#x2F;models&#x2F;claude-opus-5-5?models=claude-opus-5-high%2Cclaude-opus-5-5-high%2Cclaude-opus-5-5%2Cclaude-opus-5" rel="nofollow">https:&#x2F;&#x2F;artificialanalysis.ai&#x2F;models&#x2F;claude-opus-5-5?models=...

      Not sure about how adaptive reasoning works though as they mention adaptive reasoning for every reasoning level

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.