gpt-6IMAGE
AA benchmark questioned as GPT-6 Sol only ties Muse Spark 1.3
A post with a screenshot mocks the Artificial Analysis leaderboard, saying GPT-6 Sol only ties Muse Spark 1.3 and trails Opus 5 by three points.
Haider.
@haider1Archived 2026.09.23Originally posted 2026.09.23

Haider. @haider1Originally posted 2026.09.23
AA benchmark has to be a joke at this point no way GPT-6 sol is on par with Muse Spark 1.3 and 3 points behind opus 5
This post quotes another post: “Artificial Analysis: GPT-6 Sol and Luna halve pricing, coding agent results diverge”
x.com/haider1/status/2102477424332316727Archived 2026.09.23 10:16