Comment by gobdovan

98.6% cache hits doesn't distinguish an efficient workflow from an overly chatty linear agent repeatedly reusing the same context. Plus, it says nothing directly that the process has good useful progress per token.