Hehe me too. Yesterday. Enough is enough. Using KLM5.1 and soon deepseek
I cancelled Claude: Token issues, declining quality, and poor support
591–600 of 604 posts
Re: I cancelled Claude: Token issues, declining quality, and poor support
#592Earlier quoted context omitted.
I'm starting to think a lot of the problem people are having is just that they have unrealistic expectations. I'm not having the same problem as you and I follow a very similar methodology. I'm producing code faster and at much higher quality with a significant reduction in strain on my wrists. I doubt I'm typing that much less, but what I am typing is prose which is much more compatible with a standard QWERTY keyboa…
If the main problem is programming languages incompatibility with QWERTY, that problem has been solved many decades ago. The programmers can switch to Colemak, and save many trillions of dollars of AI expenses.
Colemak ftw
Re: I cancelled Claude: Token issues, declining quality, and poor support
#593Earlier quoted context omitted.
I'm talking about the cost of labor. If you pay someone $20,000 for labor, and they save 65 minutes worth of labor per day using a $200/mo Claude subscription, you are better off buying the Claude subscription.
I think if you (a company) pay someone for labor, your labor cannot use personal subscription and you have to pay considerably higher api prices.
You could it’s just against ToS.
But the specific numbers in my prior comments aren’t really relevant to my point. Adjust for whatever numbers you want.
Re: I cancelled Claude: Token issues, declining quality, and poor support
#594Earlier quoted context omitted.
I think if you (a company) pay someone for labor, your labor cannot use personal subscription and you have to pay considerably higher api prices.
You could it’s just against ToS. But the specific numbers in my prior comments aren’t really relevant to my point. Adjust for whatever numbers you want.
But I think they are relevant because you compare two numbers and one is much lower.
I've done some napkin math and CC code makes me more efficient when I pay 200/ month, but it wouldn't if I had to pay api prices
Re: I cancelled Claude: Token issues, declining quality, and poor support
#595Earlier quoted context omitted.
So your complaint is that they didn't pay for training data by buying every book found online? That does seem more reasonable, but makes public libraries also evil.
Except the libraries pay the fees of the books, they only serve a dedicated local region of people and by loaning a book, you will know the author of the book. For LLMs the transformative part is then removing the copyright info and serving it to you as OpenAI whatever. Sure, you can query multiple books at the same time and the technology is godlike. But the underlying issue remains. Without the original content, th…
I'm talking about the knowledge people "steal" by reading. LLMs and humans both absorb knowledge by reading. You want to tax using that knowledge that was absorbed.
It will be applied to people soon after.
Re: I cancelled Claude: Token issues, declining quality, and poor support
#596Earlier quoted context omitted.
Or just don't use AI to write code. Use it as a code reviewer assistant along with your usual test-lint development cycle. Use it to help evaluate 3rd party libraries faster. Use it to research new topics. Use it to help draft RFCs and design documents. Use it as a chat buddy when working on hard problems. I think the AI companies all stink to high heaven and the whole thing being built on copyright infringement stil…
> Or just don't use AI to write code. Anecdata, but I'm still finding CC to be absolutely outstanding at writing code. It's regularly writing systems-level code that would take me months to write by hand in hours, with minimal babysitting, basically no "specs" - just giving it coherent sane direction: like to make sure it tests things in several different ways, for several different cases, including performance, comp…
Curious to know what you are using $200/mo CC for? New applications? Business? What kind of application needs to run 2-3 clients run 24/7? How are you coming up with features. Just trying to understand the workflow.
Re: I cancelled Claude: Token issues, declining quality, and poor support
#597Earlier quoted context omitted.
> They changed it do all of the changes in a virtual cloud environment, then dump the final result at the end of the response. That’s a hallucination. All they did was hide thinking by default. Quick Google search should easily teach you how to turn it back on (I literally have it enabled in my harness).
Is anything that might be wrong or misinformation now a “hallucination”?
I’ve started co-opting it specifically in situations where someone claims something untrue that is both easy to verify and stated confidently, but also ostensibly isn’t intentionally spreading misinformation.
Re: I cancelled Claude: Token issues, declining quality, and poor support
#598Earlier quoted context omitted.
Except the libraries pay the fees of the books, they only serve a dedicated local region of people and by loaning a book, you will know the author of the book. For LLMs the transformative part is then removing the copyright info and serving it to you as OpenAI whatever. Sure, you can query multiple books at the same time and the technology is godlike. But the underlying issue remains. Without the original content, th…
I'm talking about the knowledge people "steal" by reading. LLMs and humans both absorb knowledge by reading. You want to tax using that knowledge that was absorbed. It will be applied to people soon after.
This reminds me of what happened around the time I hit year 3 in school. You could no longer buy used textbooks like everyone did from time immemorial because there was online drm making sure you had the latest textbook to take the latest quiz. I'm sure it's got even worse in the 20 years since.
Re: I cancelled Claude: Token issues, declining quality, and poor support
#599Earlier quoted context omitted.
You could it’s just against ToS. But the specific numbers in my prior comments aren’t really relevant to my point. Adjust for whatever numbers you want.
But I think they are relevant because you compare two numbers and one is much lower. I've done some napkin math and CC code makes me more efficient when I pay 200/ month, but it wouldn't if I had to pay api prices
Really? Are you using opus and letting it run for long periods? Curious as to what your workflow is.
The math is highly in favor of us using it at our company and we are paying API pricing. I don’t imagine there’s a lot of people using Claude without getting their money’s worth…?
Re: I cancelled Claude: Token issues, declining quality, and poor support
#600Earlier quoted context omitted.
But I think they are relevant because you compare two numbers and one is much lower. I've done some napkin math and CC code makes me more efficient when I pay 200/ month, but it wouldn't if I had to pay api prices
Really? Are you using opus and letting it run for long periods? Curious as to what your workflow is. The math is highly in favor of us using it at our company and we are paying API pricing. I don’t imagine there’s a lot of people using Claude without getting their money’s worth…?
Yes, recently I've been working on some research/ optimizatiom problem.
I would start claude in Yolo mode, tell it keep trying new ideas until it runs out of 1m context. (Every day I am giving it a hint to explore different directions as the sessions before)
Twice a day for a month, fits well into CC max plan.
I guess if I had to pay per token I would still use it but only for tasks where the value is clearer and immediate.