Quite scared by the fact that the original issue pointing out the actual root cause of the issue has been 'Closed as not planned' by Anthropic.

https://github.com/anthropics/claude-code/issues/46829

▲

hrimfaxi 8 hours ago | parent | next [-]

The response doesn't even make sense and appears to be written by AI.

> The March 6 change makes Claude Code cheaper, not more expensive. 1h TTL for every request could cost more, not less

Feels very AI. > Restore 1h as the default / expose as configurable? 1h everywhere would increase total cost given the request mix, so we're not planning a global toggle.

They won't show a toggle because it will increase costs for some unknown percentage of requests?

▲

stingraycharles 8 hours ago | parent [-]

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 requests where number of cache hits > 0 (to avoid people setting this flag without intent to use it), and you solved the whole issue.

▲

zozbot234 8 hours ago | parent [-]

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

	▲	stingraycharles 6 hours ago \| parent [-]
		Aren’t those latency sensitive though?

▲

sdevonoes 8 hours ago | parent | prev | next [-]

Why scared? Like, if theit software gets bad, we stop using it.

▲

comandillos 8 hours ago | parent [-]

Maybe scared wasn't the best word... but we cannot deny Opus is a great - if not greatest - model at coding and Anthropic is the only one serving it a reasonable prices when going through their subscription model.

	▲	sdevonoes 8 hours ago \| parent \| next [-]
		Sounds like an addiction to me
	▲	cmrdporcupine 6 hours ago \| parent \| prev [-]
		I mean this is blatantly false. Codex just rolled out a $100 a month plan with higher usage and lower quotas than Claude and GPT 5.4 is more capable than Opus 4.6. At least for the systems work I do. And if you can't stomach OpenAI, GLM 5.1 is actually quite competent. About Opus 4.5 / GPT 5.2 quality.

▲

rvz 8 hours ago | parent | prev [-]

When a casino is making a lot of money from gamblers, they don't care about their customers losing money, given the machines are rigged against you.

Anthropic sells you 'knowledge' in the form of 'tokens' and you spend money rolling the dice, spinning the roulette wheels and inserting coins for another try. They later add limits and dumb down the model (which are their gambling machines) of their knowledge for you to pay for the wrong answers.

Once you hit your limit or Anthropic changes the usage limits, they don't care and halt your usage for a while.

If you don't like any of that, just save your money and use local LLMs instead.