the weekly cap nerfed claude code way harder than people think, i have the numbers
retained text
# [](https://www.reddit.com/r/Anthropic/?f=flair_name%3A%22Complaint%3Asnoo_biblethump%3A%22)basically title. i run a bunch of agents on claude code all day (my whole setup lives on it) and since early october i've been logging every single request, weighting tokens by what they actually cost. not vibes, actual logs. until nov 1 the only limit that mattered was the 5 hour window. you hit it, you wait, it refills, you keep going. you could legit plan your day around it. then on nov 2 a new "weekly · all models" bar showed up in my usage panel. mid week, no heads up that i saw. so here's what my numbers say, same kind of work, same models: \- average october week: \~780M (cost weighted) \- my biggest week, mid october: \~1.5B \- the new weekly cap: \~400M that's about HALF a normal october week and a QUARTER of my peak. and the funny part is i barely touched fable in october, so it's not like i was burning the expensive model and got caught. nothing changed per token. the ceiling did. before you had a tank that refilled every 5 hours, now you have one wallet for the whole week and you can still empty it just as fast. one heavy day and half your week is gone. what i changed: \- haiku/sonnet for anything that's lookups or reading files, opus only when it actually matters \- shorter chats, because context bloat eats the week quietly \- heavy stuff gets planned for days i know i can afford it still the best tool i've used, not gonna pretend otherwise. I'm using the $125 teams seats (6.25x usage) but it went from virtually unlimited (it legit was tbf) to being able to do half the work of a normal week, crazy drop anyone else actually measuring this? curious if your ratio looks like mine or if i'm the outlier
document fetched 2026-11-19 00:41 · canonical https://reddit.com/r/ClaudeCode/comments/1xziv1a/the_weekly_cap_nerfed_claude_code_way_harder_than