The no-reasoning version scores 35% while the low reasoning one scores 17%? What... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

XCSme 20 days ago | parent | context | favorite | on: GPT-6 Astra

The no-reasoning version scores 35% while the low reasoning one scores 17%? What?

dudeinhawaii 20 days ago | [–]

I suspect this is "no reasoning set" which might be "default: medium" or perhaps some smart routing. I don't think it's literally "no reasoning".

XCSme 20 days ago | | [–]

It says (none), which usually means reasoning disabled.

I would be surprised if none = use default reasoning

silver_sun 20 days ago | [–]

It's simulating the Dunning-Kruger effect.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact