GPT-6 Astra and Fable 5.1 reportedly yield five more improvements in repeat experiment
A user says they kept the same setup and ran both models at Max effort. They interpret the result as a roughly tenfold increase in model intelligence over nearly six months.
TLDR
On September 20, 2026, a user reported repeating an experiment almost exactly six months later, using the same setup but with GPT-6 Astra and Fable 5.1, both at Max effort. They reported five improvements on top of an earlier improvement and claimed each successive gain is exponentially harder. Their conclusion: the models are roughly 10 times smarter than six months earlier.
Combined views
—
1 Source, first seen ago
— likes— comments— saves— reposts
