Skip to content

OpenAIFrontier

GPT-6.2 Sol

OpenAI's everyday model, replacing GPT-6.1 Sol at the same $2 in / $10 out per million tokens. Scores 57 on the Artificial Analysis index, against 54.6 for GPT-6.1 Sol.

Announced
Access
Public

Benchmark scores

TestScoreBest on the trackerHow it was measured
AA Intelligence Index v5about ten tests in one score57 pointsClaude Fable 5.2 64.6 pointsIndependentv5.0.1; listed as GPT-6.2 Sol (Max)Run 17 Nov 2026

Compare every model on the same tests →

Release story

Sources