3 ms·

Can we stop pretending that Mythos is a good model yet?

• sixothree 4mo ago
  What was your particular experience with it?
  • nullbio 4mo ago
    Hallucination city, doing whatever it feels like and far more than what I've bargained for, and performance on par with Opus 4.8 for coding in a large production codebase. I still have far more success with GPT 5.5, it actually follows my instructions and doesn't try to automate my entire job, which allows me to build skills and pipelines around the things I actually want automated.
    • sixothree 4mo ago
      Interesting. I only used Fable. And not for very much time. But one thing I did notice was improved adherence. Maybe it was just improved adherence to the claude.md and the instructions from the skills in use that I was noticing.