Live data from Hacker News

Claude Code weekly rate limits

news.ycombinator.com

291–300 of 724 posts

Re: Claude Code weekly rate limits

#291
post #136

That's fine, please make it VERY CLEAR how much of my limit is left, and how much i've used.

Seriously. I still find it ridiculous that even after they upped Opus' limit from 60% to 80% they don't show usage % below that. It's sapping my ability to use it quickly on the 5x plan.

Re: Claude Code weekly rate limits

#292
I understand, but I also will not pay a subscription fee for limited service. I canceled as soon as I got this e-mail. Too bad I signed up for an annual subscription last month.

This is also exactly why I feel this industry is sitting atop a massive bubble.

Re: Claude Code weekly rate limits

#293
post #229

> affecting less than 5% of users based on current usage patterns. How about adding ToS clause to prevent abuse? wouldn't that be better than having a statement with negative effect on the rest of 95%?

That just gives your team another thing to do, trying to police a ToS clause. And one would have to define abuse somehow, which is quite tricky. There are a lot of guite legitimate but expensive ways to use such a general thing as LLM compute.

Re: Claude Code weekly rate limits

#294

I'm probably not going to hit the weekly limit, but it makes me nervous that the limit is weekly as opposed to every 36 hours or something. If I do hit the limit, that's it for the entire week —a long time to be without a tool I've grown accustomed to! I feel like someone is going to reply that I'm too reliant on Claude or something. Maybe that's true, but I'd feel the same about the prospect of loosing ripgrep for a…

pay for another account to double your limit.

Re: Claude Code weekly rate limits

#295

Earlier quoted context omitted.

You can set cash on fire if you want to.

you spend 400$ per month on api usage, but your AI builds the next unicorn worth billions, where is the problem?

For $2 you can have a lottery ticket that will win you a quater of a billion dollars.

Re: Claude Code weekly rate limits

#297
post #226

Earlier quoted context omitted.

if it affects only a minority of accounts, why not figure out how to special case them without affecting everyone else is the primary question I would ask myself if I worked on this the principle: let's protect against outliers without rocking the behavior of the majority, not at this stage of PMF and market discovery i'd also project out just how much the compute would cost for the outlier cohort - are we talking $5…

> if it affects only a minority of accounts, why not figure out how to special case them without affecting everyone else that's exactly what they have done - the minority of accounts that consume many standard deviations above the mean of resources will be limited, everyone else will be unaffected.

"You're absolutely right!" i misread the announcement - thought everyone moved to primarily a weekly window but seems like 5hr window still in place and they're putting in place another granularity level control that DS teams will adjust to cutoff mostly bad actors.

correct me if I'm wrong, it's not like we have visibility into the token limit logic, even on the 5hr window?

Re: Claude Code weekly rate limits

#299
post #47

Earlier quoted context omitted.

No, it's not. It's the all you can eat buffet saying that 95% can eat all they want, but the 5% that keep sneaking food in their backpack to eat when they get can't do that anymore.

There's no lobster rolls stuffed into a backpack here. It's using the service as it was pitched, an all-you-can-eat buffer of API calls. Anything that limits what that means is scaling back access to that buffet. If the new limits are anything less than 24 * 7 / 5 times the previous limits then power users are getting shafted (which understandably is the point of this). What's worse with this model is that a runaway…

It was never unlimited. The 5 hourly limit was there from before.

Re: Claude Code weekly rate limits

#300

Earlier quoted context omitted.

You can set cash on fire if you want to.

you spend 400$ per month on api usage, but your AI builds the next unicorn worth billions, where is the problem?

You might have better odds buying scratchers - assuming you even have $400/m to invest in an enterprise without cash flow
Post reply on HN