Comment by kittoes
10 hours ago
Did you explicitly tell it to use Sonnet or Opus subagents and stick at or below high effort? Asking because such practices make a huge difference in the quality of output and the amount of tokens burned. I used one of my accounts to explore ultramax and it was just a token hog that might be worse than Opus.
I had it on whatever the recommended settings was, but maybe I should have told it to use Sonnet for most subtasks.
Even so, I'm just not that impressed, I felt like I got more done by just using Opus.
You can write a skill (many have) to lead it in how to use different models and efforts for subtasks. For searching the code base, for example, I have it use Haiku, which is fast.