GPT-6.2 Astra beats Claude Fable 5.2 by 5.5 points at just 5% of the cost

Yuchenj_UW · x · 2026-11-18

Citing OpenAI's data, GPT-6.2 Astra scores 5.5 points above Claude Fable 5.2 on Agents' Last Exam v1.1 at just 5% of the cost, and 5.2 points above GPT-6 Astra at less than a third of the cost. The accompanying score-vs-cost Pareto chart shows Astra topping out near 78%, while Claude Fable 5.2 sits around 73% at roughly $60 per task.

Related event: GPT-6.2 Astra Tops Agents' Last Exam Benchmark(2 posts)→

Original post →

More from Models

Models channel →