Yeyy just getting started

4.8 on max effort is very self conscious on the scope and where it stands no idea why you had a different impression of 5.5 at any moment of time

Also I believe this is the reason why Claude models lose to gpt recent models on long time tasks like math and so on, maybe it's a Claude code harness quirk. Opus 4.7 exits very prematurely, it…

I have an idea for that lolll, but for FX (n,p)-polygraph representation of Terms, constructed the way that you can walk in any direction (between cells) and typecheck, so parallelizable and also…