AI Reports · Epoch AI · 2026-11-17 · Blog / commentary
How good are Fable 5.2 and GPT-6.2 Astra at real work?

Epoch AI tested Fable 5.2, GPT-6.2 Astra, and four other models on an updated version of its Automation Reports benchmark, which is built from its own work tasks. No model met the standard expected of Epoch employees, though Fable 5.2 and Opus 5.5 reached contractor level on AI data center analysis. GPT-6.2 Astra averaged 65%, 17 points below Fable 5.2, despite a roughly tied ECI score.