This Friday CC wrote wall off gibberish text for me. No reason, happened twice with different gibberish text https://ibb.co/4wcVQG5k
maybe numerics issues after quantization? Looks like it really went off the rails
Anthropic downgraded cache TTL on March 6th
291–300 of 447 posts
Re: Anthropic downgraded cache TTL on March 6th
#292I also noticed this, just resuming something eats up your entire session. The past two weeks also felt like a substantial downgrade and made me regret renewing my subscription, it sucks because I wish I kept my Codex subscription instead and renewed that.
Are you locked into your current subscription?
Re: Anthropic downgraded cache TTL on March 6th
#293Earlier quoted context omitted.
> It's all circumstantial but everything points towards "desperately trying to cut costs". I have been wondering if it's more geared at reducing resource usage, given that at the moment there's a known constraint on AI datacenter expansion capability. Perhaps they are struggling to meet demand?
I wish they would just rip the bandaid to stop everybody's entitled whining. "We're sorry, what we were able to give you for $100/mo before now needs to be $200/mo (or more). We miscalculated/we were too generous/gave too much away for too little. It's a new technology, we are seeing a ton of demand, we are trying to run a business, hope you understand. If you don't want it, don't pay for it."
Are we at the point where there are external constraints that cash can't solve?
Re: Anthropic downgraded cache TTL on March 6th
#294I noticed another limitation: "An image in the conversation exceeds the dimension limit for many-image requests (2000px). Start a new session with fewer images." So I can't continue my claude code session I started yesterday.
makes sense, “a picture is worth a thousand tokens” as they say. They probably lowered the limit due to capacity issues.
Re: Anthropic downgraded cache TTL on March 6th
#295Earlier quoted context omitted.
> Perhaps Anthropic is struggling to meet demand? Yes, definitely, they’re gracefully failing to meet demand. They could also deny new customers, but it would probably be bad for business.
I once decided to deny new customers in order to be able to service current demand at the quality we wanted. It backfired and made people want our product even more. Our phones were blowing up. That approach can have unintended consequences!
You unintentionally used a common sales tactic; by decreasing supply you increase demand.
Re: Anthropic downgraded cache TTL on March 6th
#296Earlier quoted context omitted.
I wish they would just rip the bandaid to stop everybody's entitled whining. "We're sorry, what we were able to give you for $100/mo before now needs to be $200/mo (or more). We miscalculated/we were too generous/gave too much away for too little. It's a new technology, we are seeing a ton of demand, we are trying to run a business, hope you understand. If you don't want it, don't pay for it."
It is one thing to pay 100 a month to make calendar apps for your linkedin and birds on bicycles to get invited to talks, paying 200 HOWEVER
If we didn’t have the birds on bicycles, how would we know the models are getting better?
Re: Anthropic downgraded cache TTL on March 6th
#297Has anybody else noticed a pretty significant shift in sentiment when discussing Claude/Codex with other engineers since even just a few months ago? Specifically because of the secret/hidden nature of these changes. I keep getting the sense that people feel like they have no idea if they are getting the product that they originally paid for, or something much weaker, and this sentiment seems to be constantly spreadin…
> people feel like they have no idea if they are getting the product that they originally paid for
They do indeed get the product they originally paid for.
It's simply that they were suckers and didn't read the "fine" print of the product they bought.
The label says "more tokens than the lower tier".
Re: Anthropic downgraded cache TTL on March 6th
#298Earlier quoted context omitted.
> It's all circumstantial but everything points towards "desperately trying to cut costs". I have been wondering if it's more geared at reducing resource usage, given that at the moment there's a known constraint on AI datacenter expansion capability. Perhaps they are struggling to meet demand?
I wish they would just rip the bandaid to stop everybody's entitled whining. "We're sorry, what we were able to give you for $100/mo before now needs to be $200/mo (or more). We miscalculated/we were too generous/gave too much away for too little. It's a new technology, we are seeing a ton of demand, we are trying to run a business, hope you understand. If you don't want it, don't pay for it."
Just put everyone on pay per use with the API and rip the band aid off.
Re: Anthropic downgraded cache TTL on March 6th
#299Earlier quoted context omitted.
To be clear they weren’t banned from Claude usage, they were required to use the API and API rates rather than Claude Max tokens. Claude code uses a bunch if best practices to maximize cache hit rate. Third party harnesses are hit or miss, so often use a lot more tokens for the same task.
nah this doesn't explain it. most of the users of those third party harnesses care just as much about hitting cache and getting more usage.
Sure. The question is whether they have the same level of expertise and prioritization that Anthropic does.
Re: Anthropic downgraded cache TTL on March 6th
#300Earlier quoted context omitted.
claude -p not working would be instant unsubscribe downgrade from Max to Pro and further drive my use of codex. I use both but overall have noticed I reach for Claude less than codex lately because claude keeps getting slower and slower (I have not noticed a drop off in quality, but I use it less and less so maybe I'm not in a good position to notice). Generally I find codex and claude make a good team. I'm not a hea…
I might consider switching to codex from claude pro 20x but I need the post tool use, pre file write and post user message hooks. Waiting on codex to deliver. - pre file write -> block editing code files without a task and plan of work - post tool use -> show next open checkbox in the task to the agent, like an instruction pointer - post user message -> log all user messages for periodic review of intent alignment Th…
Why couldn’t you use Claude code harness with codex? The requests can be proxied to OpenAI.