Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

github.com

Post reply on HN
Print view View original

311–320 of 695 posts

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#311

Post by ramon156 »

bcherny wrote: post #210

Hey all, Boris from the Claude Code team here. We've been investigating these reports, and a few of the top issues we've found are: 1. Prompt cache misses when using 1M token context window are expensive. Since Claude Code uses a 1 hour prompt cache window for the main agent, if you leave your computer for over an hour then continue a stale session, it's often a full cache miss. To improve this, we have shipped a few…

Boris, wasnt this the same thing ~2 weeks ago? Is it the same cache misses as before? What's the expected time till solved? Seems like its taking a while

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#312

Post by HauntingPin »

siliconc0w wrote: post #167

Switched back to codex for the promotion. Opus at the start of the year was GOAT- just relentless at chewing through hard problems. Now it spins on pretty easy work (took three swings just to edit a ts file) and my session is like 1-3 prompts (downgraded to the $20 plan but still)

Had a single prompt the other day where it just tried to examine dependencies that weren't relevant until it hit the rate limit. That was my first prompt of the day. On a task that it was able to do quickly and successfully many times before.

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#313

Post by Jensson »

hk__2 wrote: post #280
ryandrake wrote: post #171

Earlier quoted context omitted.

Yea, I found myself maxing out the $20/mo plan occasionally, so I tried the $100/mo, but I don't think I even once even approached the session limit, let alone the weekly limit. And this is doing what I would consider heavy, continuous programming. I probably ought to go back down to $20 one. It would be nice if they had a cheaper tier in between them, but the tiers they have are probably a good business trick to get…

I’m on the $20/mo plan right now and I hit the limit in under an hour, sometimes 20-30 minutes. I don’t understand how people can work with this plan; maybe it was better before?

Probably depends on the size of your codebase and how AI friendly it is.

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#314

Post by stingraycharles »

zozbot234 wrote: post #71
stingraycharles wrote: post #57

Earlier quoted context omitted.

Sounds like a decision I would make when memory is expensive and you want to get rid of the very long (in time) tail of waiting 1h to evict cache when a session has stopped. There must be a better way to do this. The consumer option is the pricing difference. If they’d make cache writes the same price as regular writes, that would solve the whole problem. If you really want to push it, use that pricing only for reque…

Memory is expensive? If reads are as rare as they claim you can just stash the KV-cache on spinning disk.

Aren’t those latency sensitive though?

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#315

Post by HauntingPin »

brookst wrote: post #193
nickstinemates wrote: post #181

It feels so weird to me - people are exhausting their quotas while I am trying very hard to even reach mine with the $200 plan. We're generating all of the code for swamp[1] with AI. We review all of that generated code with AI (this is done with the anthropic API.) Every part of our SDLC is pure AI + compute. Many feature requests every day. Bug fixes, etc. Never hit the quota once. Something weird is definitely goi…

My hypothesis is that people who have continuous sessions that keep the cache valid see the behavior you’re describing: at 95% cache hits (or thereabouts), the max plan goes a long way. But people who go > 5 minutes between prompts and see no cache, usage is eaten up quickly. Especially passing in hundreds of thousands of tokens of conversation history. I know my quote goes a lot further when I sit down and keep sess…

Why is it our job to micromanage all this when it used to work fine without? Something's clearly changed for the worse. Why are people insisting on pushing the responsibility on paying users?

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#316

Post by eastbound »

bcherny wrote: post #253
yummytummy wrote: post #249

Earlier quoted context omitted.

Ah, so cache usage impacts rate limits. There goes the ”other harnesses aren’t utilizing the cache as efficiently” argument.

Claude Code is the most prompt cache-efficient harness, I think. The issue is more that the larger the context window, the higher the cost of a cache miss.

I’m sorry but when you wake up in the morning with 12% of your session used, saying “it’s the cache” is not an appropriate answer.

And I’m using Claude on a small module in my project, the automations that read more to take up more context are a scam.

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#317

Post by gruez »

ImPostingOnHN wrote: post #289
gruez wrote: post #169

Earlier quoted context omitted.

>and business expect the level of productivity witnessed before, will have no choice but cough up whatever providers bill us. Is that bad? After all, even if they hiked to price infinity, you wouldn't worse off than if AI didn't exist because you could still code by hand. Moreover if it's really in a "business" (employment?) context, the tools should be provided by your employer, not least for compliance/security rea…

> if they hiked to price infinity, you wouldn't worse off than if AI didn't exist because you could still code by hand This was addressed by the words that you perhaps mistakenly omitted from your quote: > Once people won't be able to think anymore... People who aren't able to think anymore, can't still code by hand. Think "Idiocracy".

>People who aren't able to think anymore,

OpenAI and Anthropic have been getting stingy with their plans and it's only it's been what, 1 year, maybe 2 since vibecoding was widely used in a professional context (ie. not just hacking together a MVP for a SaaS side hustle in a weekend)? I doubt people are going to lose their ability to think in that timespan.

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#318

Post by amirhirsch »

Ucalegon wrote: post #301
bcherny wrote: post #293

Earlier quoted context omitted.

We are taking it seriously, and are continuing to investigate. We are not trusting the metrics.

[flagged]

Dude is on hacker news on a Sunday. half the GDP of the world is competing with him. What metrics would you like to see?

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#319

Post by HauntingPin »

postalcoder wrote: post #14

I had used Claude Code max as my daily driver last year and this sort of drama was par for the course. It's why I migrated entirely to Codex, despite liking Claude, the harness, more. There's this honeymoon period with Claude you experience for a month or two followed by a trough of disillusionment, and then a rebound after a model update (rinse and repeat). It doesn't help that Anthropic is experiencing a vicious co…

This past week was a nightmare in trying to get Claude to do any useful work. I've cancelled my subscription and everybody else here having problems should too. I don't think Anthropic cares about anything else.

Re: Pro Max 5x quota exhausted in 1.5 hours despite moderate usage

#320

Post by rafaelmn »

meetingthrower wrote: post #56

I don't get it. Last week on the 100 bucks plan I generated probably 50k LOC (not a quality measure for sure!) and just barely kissed the weekly limit. I did get rate limited on some sessions for sure, but that's to be expected. I'm curious what are people doing that is consuming your limits? I can't imagine filling the $200 a month plan unless I was essentially using Claude code itself as the api to mass process stu…

I'm in the same camp but I mostly do backed. My coworker doing frontend is chewing through rate limits consistently. React code is quite logic shallow, stuff gets pulled in all over so not localized, especially when you start using js styling frameworks - hundreds of k of tokens to do simple changes.

If you start to parallelize and you have permission prompts on you're likely missing cache windows as well.

Post reply on HN