So many negative comments here. Among my circle, it’s not uncommon for people to have multiple $200 accounts. This seems like it would make things easier for them by letting them use one account instead of juggling two or more.
What they’re doing with billions of tokens every week, beats me. But these users absolutely do exist.
For less than the cost of this plan, you can get ChatGPT Pro 200, Claude Max 5x, and Kimi Code. At least that's what I'm doing, and the model diversity does help with adversarial code review loops.
At this point, why even bother with subscriptions and other packaged products of these model providers anyways. It’s just an extra hassle of juggling between them plus it introduces additional data privacy concerns.
Do things in a sustainable fashion and treat these models like the swappable commodity that they are by using them via abstraction layers. Bonus point is that you might be surprised how much more you can get for your money by intelligently optimizing specific model usage.
This approach is fun and rewarding but I think it makes sense to do this if/when models really stall out. Right now opus 5.5 is... amazing. I wouldn't use anything else.
Subscriptions are still subsidized over API, so just like Uber I'll continue to milk them until I can't anymore (Uber back in the 2010s was incredible, you could get damn near anywhere for less than 5 bucks or so).
The drivers are still contractors and get very little of the ride money unless demand is high. Uber has been very successful at legislating carveouts for rideshare drivers to protect its margins.
“Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent.”
Your concern is legit. Pro is a personal plan under OpenAI's consumer privacy policy: "When you use our services for individuals such as ChatGPT, Sora, or Operator, we may use your content to train our models." As for the API it goes under the business one: "By default, we do not use your business data for training our models."
Claude has the same personal-professional split. Free, pro and Max have trained by default since Anthropic's policy change in August 2025. API and team plans didn't. Choosing between one and another is also choosing a training default. The subscriptions offers the opt-out setting, Data Controls for ChatGPT and Account settings for Claude, and it's worth checking before paying.
Yikes. I'm not the poster you're responding to, but
1. In this absolutely wild era of model progress, with frontier and near frontier models leapfrogging themselves seemingly on a weekly basis it's good to get experience with a variety of providers
2. This isn't exactly hard with OpenRouter!
3. I've had very good results having models from one provider critique plans and code written by models from another provider. Can't do that without subs to at least two providers
Can you please explain to me how the announcement is anything but a negative? Especially for such power users. That same 200$ plan now allows for half the usage it did when I subscribed a couple weeks ago. Tibo and co can dress this pig up however they want, it's still less usage than I bargained for.
>What they’re doing with billions of tokens every week, beats me.
What intrigues me more is how are they not done building whatever they were building? LLMs have been jaw-droppingly good for a good 6 months now.
The people spending thousands on LLM bills a month should've been done building significant chunks of Microsoft's or Google's software portfolio by now.
I guess what I mean is how is producing code still the limiting factor in whatever they're doing?
I produce a firehose of personal software and I still cannot max out a $20 plan.
There's nothing wrong with making really simple small scripts. I do that as well; I'm sure everyone does. However, imo for anything ambitious, $20 buys you about a day give or take before your weekly quota gets eaten. I've been able to burn that in less than a day with about 2-3 sessions. (No, my projects aren't even close to Azure in scale.) To avoid that, you would have to do a lot of optimization like having your frontier subcription orchestrate openrouter agents to do the actual work. I'm not even sure that's possible because my $20 sub burned with just analysis and not implementation AND it was using a lot of tools to offload deterministic work.
Yeah… i have multiple $200 Claude accounts and a supergrok heavy account and also a max $500. The max $500 was a new addition today, but i have my whole company basically wired and ran by ai with a central HQ setup. While the $1k month price is high, that $1k gives me the equivalent of a full dev team working 24/7. Pretty soon we’ll need more horsepower and will move to a model like kimi 3 or local hosting our own model. Thats something i still have to learn more about. But yeah, build a system where you can swap any model or ai platform in and out easily
Would you ask the same thing if they happened to employ a single quality software developer? Can you imagine paying said developer 12k a year and expect them to deliver? How about a personal tutor who you have access to 24/?
I've been dropping the equivalent of a car payment for three months and it's been worth every single penny; even if I never actually sell a product.
It was originally a small project in scope, just a basic GameBoy emulator, a shader VM, and some Azure integrations. Just a bunch of things I wanted to dabble with in order to learn and have fun. Over the past few months it has grown immensely into something resembling an operating system more than a toy engine. There's even a custom DSL for authoring worlds that abstracts over the state/rendering models; it's completely ridiculous!
Will I make any money off the thing? Probably not, don't care. The things I've learned along the way have been worth way more than the amount of money I burned on a few monthly subscriptions. Learning how to interact with agents and treat them like a team was worth it alone; the product I have now is simply a nice bonus.
> What they’re doing with billions of tokens every week, beats me. But these users absolutely do exist.
I think you can just become slightly careless and off it goes. I just have a £20/mo ChatGPT and I'm getting it to do some HomeLab Kubernetes YAML for me that I have no desire to hand-type. I asked it a question and it zoomed off looking at all sorts of websites, running various git commands, doing anything except asking me for an answer to a question.
It ran through about 60% of my 5-hour allocation in those few minutes, which was an unheard-of rate for me before that, and I can imagine if I had zillions of tokens available I just wouldn't have noticed. As it is, my AGENTS.md now exists and is fairly stern.
Latest models have a much better sense of time and token budget, so if you instruct it for a quick answer or to not burn many tokens it will try to respect that
azuanrb · · focus · HN ↗
What they’re doing with billions of tokens every week, beats me. But these users absolutely do exist.
alasano · · focus · HN ↗
If it was a 50x $500 plan it would be a different story.
bezier-curve · · focus · HN ↗
alasano · · focus · HN ↗
6.1 will probably perform even better.
Opus 5.5 is better in that regard than Fable 5.1 except for some bugs with more complexity which it was uniquely suited to.
purrry · · focus · HN ↗
Do things in a sustainable fashion and treat these models like the swappable commodity that they are by using them via abstraction layers. Bonus point is that you might be surprised how much more you can get for your money by intelligently optimizing specific model usage.
timcobb · · focus · HN ↗
satvikpendem · · focus · HN ↗
robertlagrant · · focus · HN ↗
Back then the trips were subsidised by VC money, and also the drivers were contractors, so far lower internal costs for Uber.
satvikpendem · · focus · HN ↗
SR2Z · · focus · HN ↗
chaostheory · · focus · HN ↗
“Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent.”
UltraSane · · focus · HN ↗
tux404 · · focus · HN ↗
Claude has the same personal-professional split. Free, pro and Max have trained by default since Anthropic's policy change in August 2025. API and team plans didn't. Choosing between one and another is also choosing a training default. The subscriptions offers the opt-out setting, Data Controls for ChatGPT and Account settings for Claude, and it's worth checking before paying.
MuffinFlavored · · focus · HN ↗
Be a better friend to yourself. Value your time more.
JohnBooty · · focus · HN ↗
1. In this absolutely wild era of model progress, with frontier and near frontier models leapfrogging themselves seemingly on a weekly basis it's good to get experience with a variety of providers
2. This isn't exactly hard with OpenRouter!
3. I've had very good results having models from one provider critique plans and code written by models from another provider. Can't do that without subs to at least two providers
kittoes · · focus · HN ↗
roncesvalles · · focus · HN ↗
What intrigues me more is how are they not done building whatever they were building? LLMs have been jaw-droppingly good for a good 6 months now.
The people spending thousands on LLM bills a month should've been done building significant chunks of Microsoft's or Google's software portfolio by now.
I guess what I mean is how is producing code still the limiting factor in whatever they're doing?
I produce a firehose of personal software and I still cannot max out a $20 plan.
kittoes · · focus · HN ↗
Scope creep? When you give a mouse a cookie and all that...
nujabe · · focus · HN ↗
mapontosevenths · · focus · HN ↗
satvikpendem · · focus · HN ↗
chaostheory · · focus · HN ↗
LoadbearingGun · · focus · HN ↗
chaostheory · · focus · HN ↗
Keyframe · · focus · HN ↗
CamperBob2 · · focus · HN ↗
bushbaba · · focus · HN ↗
jbfromkc · · focus · HN ↗
nixgeek · · focus · HN ↗
If it's $1000 per month to make $0 because you haven't found PMF, less obviously a good deal unless you have a pathway to revenue?
devrwoody · · focus · HN ↗
Natfan · · focus · HN ↗
kittoes · · focus · HN ↗
I've been dropping the equivalent of a car payment for three months and it's been worth every single penny; even if I never actually sell a product.
Natfan · · focus · HN ↗
kittoes · · focus · HN ↗
<a href="https://github.com/ByteTerrace/Puck" rel="nofollow">https://github.com/ByteTerrace/Puck
It was originally a small project in scope, just a basic GameBoy emulator, a shader VM, and some Azure integrations. Just a bunch of things I wanted to dabble with in order to learn and have fun. Over the past few months it has grown immensely into something resembling an operating system more than a toy engine. There's even a custom DSL for authoring worlds that abstracts over the state/rendering models; it's completely ridiculous!
Will I make any money off the thing? Probably not, don't care. The things I've learned along the way have been worth way more than the amount of money I burned on a few monthly subscriptions. Learning how to interact with agents and treat them like a team was worth it alone; the product I have now is simply a nice bonus.
robertlagrant · · focus · HN ↗
I think you can just become slightly careless and off it goes. I just have a £20/mo ChatGPT and I'm getting it to do some HomeLab Kubernetes YAML for me that I have no desire to hand-type. I asked it a question and it zoomed off looking at all sorts of websites, running various git commands, doing anything except asking me for an answer to a question.
It ran through about 60% of my 5-hour allocation in those few minutes, which was an unheard-of rate for me before that, and I can imagine if I had zillions of tokens available I just wouldn't have noticed. As it is, my AGENTS.md now exists and is fairly stern.
wordpad · · focus · HN ↗
herval · · focus · HN ↗