GPT-6 Sol leads AutomationBench over Fable 5.1 at 88% lower cost per task

OpenAIDevs · x · 2026-09-23

OpenAI shared AutomationBench results (measuring whether models can complete real business workflows across apps): GPT-6 Sol (xhigh) leads Fable 5.1 (max with Opus 5 fallback) at 88% lower reported cost per task. It also cites DeepSWE v1.1: GPT-6 Sol (max) nearly matches Claude Fable 5 (xhigh) at 80% lower cost per task, and Luna (max) matches Fable 5 (medium) at 96% lower cost.

Related event: OpenAI Launches GPT-6 Sol and Luna at Half the Price(48 posts)→

Original post →

More from Models

Models channel →