logoalt Hacker News

XCSme • yesterday at 7:06 PM • 1 reply • view on HN

Yup, in my benchmarks usually medium or high do better than xhigh or max


Replies

spider-mario • today at 12:03 AM

You’re talking about something else (reasoning levels). The parent comment is referring to e.g. GPT Terra as opposed to Luna, Sol or Astra, Claude Sonnet as opposed to Haiku, Opus or Fable, or Gemini Flash/Hydrogen as opposed to Flash-Lite, Argon or Carbon.

➕ show 1 reply