i had a good several week run with gpt 5.5 today was completely miserable im so mad rn

@thdxr I dunno what’s happened the last few days or so but it’s been some of the worst experiences with LLMs Either I’m working on something that is non trivial enough for it to start hallucinating, or they’re scaling back for costs.

@dreamsofcode_io @thdxr Saw your message on that 2d ago, and couldn't relate, but had a similar feel today. Can't tell if it's coming from my increased confidence in 5.5, making me expect too much in high complexity task, or if model was really tweaked