Comment by impulser_
4 hours ago
Usage is actually Claude now because of Opus 5.5 since it a better model that Astra. I maxed out my 200$ Claude plan with 10b token on Opus 5 and 5.5 is cheaper. I maxed out two Codex accounts with like not even 5b tokens.
Have you used 10b/5b tokens over the course of a week or over the course of a month?
It was 9.4B to be exact and it was over the course of two days lol. It was between two projects so 99% of them were cached reads.
The GPT was about 1B on two projects on 300$ worth of plans all on Astra and I capped out on usage.
Anthropic caching must be better because the cache rates are better on Claude models.