Pro Max 5x quota exhausted in 1.5 hours despite moderate usage • Hacker News Board

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

#691

Post by stitched2gethr » Thu, Apr 16, 2026, 2:36 AM UTC

danmaz74 wrote: post #374

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…

Could we get an option to use Opus with a smaller context window? I noticed that results get much worse way earlier than when you reach 1M tokens, and I would love to have a setting so that I could force a compaction at eg 300k tokens.

Try this: `claude --model claude-opus-4-6`

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

#692

Post by jatora » Thu, Apr 16, 2026, 11:04 PM UTC

ozozozd wrote: post #558

jatora wrote: post #365

Earlier quoted context omitted.

For reference, users on claude max 20x who hit their weekly quota would have spent roughly ~$6,000/month in the API. (Source: my own usage) So you just aren't in the same realm of usage. Maybe that is why you don't understand?

I guess I could’ve been clearer. What I don’t understand is why people aren’t trying models that are 10x and in some cases 100x cheaper. Though unclear why you’d assume all my usage would be on Claude Opus when I mentioned “a bunch of Chinese models?” Unless this is a flex about how many tokens you burned. In which case, congrats...?

Oh woops didnt see the part about Chinese models my bad

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

#693

Post by eurekin » Sat, Apr 18, 2026, 12:27 PM UTC

cdelsolar wrote: post #504

eurekin wrote: post #238

Earlier quoted context omitted.

I'm adding two extra gpus to my local rig. Turns out qwen 3.5 122b is already enough to handle (finish with moderate guidance) non-planning parts of my tasks.

what kinda gpus are you using?

3090s

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

#694

Post by HarHarVeryFunny » Sat, Apr 18, 2026, 9:58 PM UTC

bcherny wrote: post #257

mvkel wrote: post #227

Earlier quoted context omitted.

Why did this become an issue seemingly overnight when 1M context has been available for a while, and I assume prompt caching behavior hasn't changed? EDIT: prompt caching behavior -did- change! 1hr -> 5min on March 6th. I'm not sure how starting a fresh session fixes it, as it's just rebuilding everything. Why even make this available? It feels like the rules changed and the attitude from Anth is "aw I'm sorry you di…

> 1hr -> 5min on March 6th This is not accurate. The main agent typically uses a 1h cache (except for API customers, which can enable 1h but it is not on by default because it costs more). Sub-agents typically use a 5m cache.

What does it mean that sub-agents use a 5 min cache? Is this just for growing contexts submitted by subagent itself? What about fixed sub-agent context prefixes such as tool definitions? Are those 1hr TTL?

It seems that a more flexible cache control mechanism, useful for variable duration sub-agents, would be more like an arena allocator. Let the client tag their API key activity with different "cache group" (arena) identifiers, then provide an API method to let them free each cache group when they are finished with it. Each sub-agent would then use it's own cache group and clear it when the sub-agent exits, rather than just having a fixed 5min or 1hr TTL. The client could provide a default TTL for each cache group to use in case they forget to free.

Context prefixes like tool definitions that will be the same for multiple invocations of the same sub-agent type could then be created (maybe by main agent) with a different cache group, and a longer default TTL.

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

#695

Post by lw1981 » Thu, Apr 23, 2026, 3:27 PM UTC

chandureddyvari wrote: post #40

Claude has gotten noticeably worse for me too. It goes into long exploration loops for 5+ minutes even when I point it to the exact files to inspect. Then 30 minutes later I hit session limits. Three sessions like that in a day, and suddenly 25% of the weekly limit is gone. I ended up buying the $100 Codex plan. So far it has been much more generous with usage and more accurate than Claude for the kind of work I do.…

I’ve been wondering whether this is less a “model got worse” problem and more an constraints problem.

If the files and task boundary are already known, letting the system keep exploring/replanning inside one accumulating context seems like the wrong default. It makes the agent pay over and over to rediscover what it should already know, and every extra turn increases the chance it wanders into a different approach.