We were not clear that “unlimited usage” was only for Auto and not all other models
Like it was too much of an engineering challenge to display the consumption of the “included usage”? Sorry, I don’t buy that explanation. Even now that usage information is quite hidden.
We moved from request-based pricing to $20 of included usage on Pro.
It was disappointing to find that out only when I’ve almost consumed that “included usage” because I was certain that it’s unlimited and I carelessly created big contexts when AI couldn’t produce a working code, and I just carelessly pushed it to “try harder”. Previously, when I clearly knew the 500 limit, I never got even close to using it up.
I guess I’m tired of trying to figure out things with cursor:
- why quality is so inconsistent and why sometimes rules are clearly ignored,
- what happens with the context, how much is used
- and now pricing shenanigans