Reactions from ranked influencers
2 posts
In my experience gpt-5.6 is VASTLY better at fundamental Math research than Fable. Before 5.6 dropped I was working for several days with Fable on a big Math project, and it seemed like it was making some progress. Once 5.6 dropped, I used it to go over what Fable had done up to that point, and it turned out there was hardly any real progress.
why are we seeing gpt-5.6 take down a ton of open mathematics problems, but none from fable? is it just that openai is giving everyone more tokens or is that the model is better at mathematics or what
I definitely agree. 5.6 is superior to Fable at math, Pro version even more so. Still, in my tests not quite at 5.2 Pro levels - that version had just an unbelievable reasoning budget (and probably a very high n in best-of-n)
In my experience gpt-5.6 is VASTLY better at fundamental Math research than Fable. Before 5.6 dropped I was working for several days with Fable on a big Math project, and it seemed like it was making some progress. Once 5.6 dropped, I used it to go over what Fable had done up to that point, and it turned out there was hardly any real progress.