WHAT NEW IN X10.31 SAT

AI collections from X · 22 collections · 63,170 posts

gpt-6THREAD

GPT-6 Sol and Luna hallucinate less than predecessors at max effort, Artificial Analysis says

Artificial Analysis reports that both GPT-6 Sol and Luna hallucinate less than their predecessors at max effort.

Artificial Analysis
@ArtificialAnlysArchived 2026.09.23Originally posted 2026.09.23
THREADPosted in 2 parts
  1. GPT-6 Sol and Luna both hallucinate less than their predecessors at max effort.

    GPT-6 Sol and Luna hallucinate less than predecessors at max effort, Artificial Analysis says (thread part 1)
  2. GPT-6 Sol (max) gains 2 points in the Artificial Analysis Coding Agent Index at half the Cost per Task of its predecessor, while GPT-6 Luna (max) regresses by 2 points.

    GPT-6 Sol and Luna hallucinate less than predecessors at max effort, Artificial Analysis says (thread part 2)
Artificial Analysis @ArtificialAnlysOriginally posted 2026.09.23
Open original on X
x.com/artificialanlys/status/2102462968084758810Archived 2026.09.23 10:21