New subscriptions that aren’t eligible for grandfathering include a lower usage allowance than previously offered with Pro 200 to reflect our increasingly efficient models.
Everyone happily paid the way for the behemoth that squeezed everyone out of competition and then started recoup the ROI aggressively, and no one recognizes oneself as a part of the problem, just like in the current AI pandemonium.
Yeah no. $200 was the top for me. Which doesn't super matter to me, I pretty much only use luna. I can't tell the difference between luna vs sol/astra other than those larger models do more ambitious things that I don't ask for.
Yea, the $500 package is likely a rebranded $200 from a few months ago. Keep in mind all providers have been cutting token allowances for their subs for a few months now
How would that work? It's not simple to set a price, when you have a fixed asset. Seems more complicated to set price on a product (ChatGPT), where the product is just a shell for something else entirely (the model), that changes by the month.
Decent(-ish), but doesn't provide good value for the money, sadly. They still refuse to tell you exactly how much usage you're paying for. My original suggestion a few months ago was to provide a $500/$1000 tier but make it _unlimited_, which this doesn't appear to be.
Seriously, even the "Learn more about limits" link just takes you to the posted page where it literally says nothing about limits or usage. Not even x20 or x10 or whatever, just zero information on what "more usage" actually means. I don't understand why they can't provide how much token usage you get.
This day may go down in history as one of the worst in OpenAI's history. It's so bad it's funny.
OpenAI announced they built clones of apps that already exist, codex for normal people, and 2.5x price increase for the product that is already getting decimated by competition.
What is happening at OpenAI?
I can't imagine OpenAI not backtracking this unless Anthropic does the price hike too.
If your Pro 200 subscription was active at the eligibility cutoff or during the seven days before it, you’re eligible to keep your previous included usage allowance through Oct 29, 2026 while you have an active Pro 200 subscription.
After that date, your subscription will move to the lower included usage allowance. Your subscription price stays at $200/month.
Imagine buying this shit for a year, upfront, and then having to deal with this. Nice.
I wonder how this will go with the new releases like MiMo-v2.6-Pro or future GLM, DeepSeek, and Kimi releases around the corner. Once planned with Fable or Astra, the build process isn't much different with the open weight models from my usage.
you can pretty much run frontier-level models locally now with GLM 5.3 Flash and Qwen 3.8 Flash Next so this pricing feels very geared towards squeezing people who are still dependent on cloud subscription models only and don't know how to divert grunt work to cheaper models automatically with the 'smarter' frontier models acting as orchestrators and refinement
At this rate I expect end-to-end work getting done, no babysitting whatsoever. The agents should literally be calling me and talking about their problem if something needs my input.
Very cool how I'm supposed to feel good about paying $200/mo (a VERY expensive subscription!) for months, just to be told that in the near (but not immediate) future this subscription will lose half of its value and then I'll have to pay 2.5x as much to get 1.25x the value I used to get.
The performance changes between models are almost static. Recently, GPT has shifted almost entirely toward token efficiency.
Astra is good at making 3D assets, but realistically its coding was not as good as GPT 5.6 Sol.
Have we reached a point where getting better at one thing means starting to get worse at another?
This announcement is the worst one yet.
Nerfing existing usage
A more expensive top-tier plan
Showing off how slow Dots is right in the demo video
And GPT Space—are they telling us to use it instead of Notion or what...
Overall, throughout the entire announcement, everything OpenAI wants me to look forward to just leaves me, as a developer, disappointed.
Yep. It's hitting limits because models have effectively reached the logical endpoint of how "precise" their outputs can become. In order to overcome this they need to either a) get excellent at outputting and working with massive codebases (10+ million loc) since a singificant % of systems simply cannot be done in less than that (the language itself isn't expressive enough) or b) reach the next stage of intelligence where the models are somehow able to output sequences that can solve vast complex problems with a very narrow output. Option b isn't really possible (various statistical/computational fundamental limitations forbid it) and option A is _insanely_ expensive. Which is why all of the demos the frontier labs have been pushing out are either really impressive one shot demos where the model is able to take a concise input and produce something great on its own, or very long running internal sessions where they claim to do something vast with minimal oversight (NavierStokes, C compiler, Bun rewrite).
actual lol
Oh don't worry, you get to keep your higher usage allowance for one month. hahahaha
Enjoy the Lyft v Uber era while it lasts!
lol yeah, I opened the Uber app the other day to $100 fares, remembered a time when I could get anywhere in SF at any time for like $5.
$5 flat group rides, $8 flat single rides anywhere in the city. The glory days.
Everyone happily paid the way for the behemoth that squeezed everyone out of competition and then started recoup the ROI aggressively, and no one recognizes oneself as a part of the problem, just like in the current AI pandemonium.
I guess they got tired of people saying Codex plans were a better value than Claude Code.
they were! now they are not
Conveniently Opus 5.5 is actually usable vs 5.
But I'm still for investing in local models.
Tired of their investors saying that
Related:
Tomorrow we are re-opening the Pro $200 subscription
https://news.ycombinator.com/item?id=49889306
Yeah no. $200 was the top for me. Which doesn't super matter to me, I pretty much only use luna. I can't tell the difference between luna vs sol/astra other than those larger models do more ambitious things that I don't ask for.
Why would I pay $500 now when the $200 features were just rug-pulled? I mean, Pro 1000 is on the horizon at this point, no?
Yea, the $500 package is likely a rebranded $200 from a few months ago. Keep in mind all providers have been cutting token allowances for their subs for a few months now
Am I just out of touch with our times if I think hand-wavey usage limits on a 500$ subscription is crazy?
Im surprised some countries dont have consumer protection laws that would force them to be more explicit/accurate
How would that work? It's not simple to set a price, when you have a fixed asset. Seems more complicated to set price on a product (ChatGPT), where the product is just a shell for something else entirely (the model), that changes by the month.
but to do accurate billing, they would have to tell us when our usage triggers a guard rail and causes our query to be rerouted or denied.
And if they do that accurately, there's no stopping someone from mapping the guard rails.
Clearly that outcome is more uncomfortable than what we have now, which is enough people seem to be okay with vague billing.
Move fast and bill things
Yeah it's wild that they don't outright state the specific usage limits.
Just trust me bro accounting.
Remember when people talked abouthow these models are similar to gambling. Well, might be preetty much loot box logix crossed with DLC
It filter to people/organizations who have a hand-wavey relationship with $500.
Yes, you are. All of this was obvious from the start. The prices will go up way more soon (tm)
Given that I burn through 2 x $200 subscriptions, I really hope Sol 6.1 is indeed as good as Astra (￣ー￣；
Decent(-ish), but doesn't provide good value for the money, sadly. They still refuse to tell you exactly how much usage you're paying for. My original suggestion a few months ago was to provide a $500/$1000 tier but make it _unlimited_, which this doesn't appear to be.
Lowering usage limits, increasing price.
This is just the beginning. Migrate to open models ASAP.
For a second I was worried I somehow missed 490 versions.
What a strange idea to name your product after it's price.
Horrid value.
Previous $200 tier: 20x pro value
New $200 tier: 10x
New $500 tier: 25x, plus ultrafast mode but it drains 6x the cost (!!)
Where do you see that? I can't even see the $500 plan on their pricing page.
The pricing is on the linked page but does not appear to be selectable in ChatGPT UI yet.
The $500 25x was mentioned by Altman on the livestream
The $200 cut to 10x was announced only formally on Twitter
However none of their product pages mention concrete #s, so I guess value will only degrade more over time.
I checked it's actually awful lol
The plan page just says "More usage" it doesn't say any 10x multiplier. Literally just says "More" than plus.
Seriously, even the "Learn more about limits" link just takes you to the posted page where it literally says nothing about limits or usage. Not even x20 or x10 or whatever, just zero information on what "more usage" actually means. I don't understand why they can't provide how much token usage you get.
This day may go down in history as one of the worst in OpenAI's history. It's so bad it's funny.
OpenAI announced they built clones of apps that already exist, codex for normal people, and 2.5x price increase for the product that is already getting decimated by competition.
What is happening at OpenAI?
I can't imagine OpenAI not backtracking this unless Anthropic does the price hike too.
The $20/month subscription is an excellent deal. I wonder what these whales are doing with all those tokens?
I mean, other than Yegge. He seems like a pretty extreme outlier?
Found a whale: "It took OpenAI's agents 130 billion tokens to crack a 90-year-old math problem" https://www.businessinsider.com/openai-math-problem-solved-t...
It's funny how, just before that, they tactfully cut the limits on the $200 subscription.
Been doing it slowly for months week over week allowances have been getting cut by 5-10% consistently
Imagine buying this shit for a year, upfront, and then having to deal with this. Nice.
No actual mention of the actual 'cutoff date' anywhere to be seen. Was it today? Yesterday? 6 months ago?
I wonder how this will go with the new releases like MiMo-v2.6-Pro or future GLM, DeepSeek, and Kimi releases around the corner. Once planned with Fable or Astra, the build process isn't much different with the open weight models from my usage.
you can pretty much run frontier-level models locally now with GLM 5.3 Flash and Qwen 3.8 Flash Next so this pricing feels very geared towards squeezing people who are still dependent on cloud subscription models only and don't know how to divert grunt work to cheaper models automatically with the 'smarter' frontier models acting as orchestrators and refinement
https://this.os.isfine.org/blog/posts/chinese-flash-models-a...
Great naming scheme that doesn’t even survive exchange rates, or future price changes.
At this rate I expect end-to-end work getting done, no babysitting whatsoever. The agents should literally be calling me and talking about their problem if something needs my input.
We’re getting pretty close to that being the normal process at this point
I know people who have been at this as stage since around 4o with their own custom systems
Me too, I'd say we are about two weeks away
Very cool how I'm supposed to feel good about paying $200/mo (a VERY expensive subscription!) for months, just to be told that in the near (but not immediate) future this subscription will lose half of its value and then I'll have to pay 2.5x as much to get 1.25x the value I used to get.
Fuck that shit!
People complained about burning their weekly usage in a day on Pro 200, so OpenAI's response to that is an even more expensive plan?
Is this Astra running on Cerebras?
No this is just the 500$ plan. They offer "ultrafast" now which is ~300tps which is 8x the speed for 6x the price.
I think 300 tps might be roughly what an astra sized model would be at if running on Cerebras hardware no?
Or are they simply lower the batch sizes on whatever hardware they are currently running?
Venture capital doesn't want to pay for our tokens anymore, it seems
Is AI development hitting its limits?
The performance changes between models are almost static. Recently, GPT has shifted almost entirely toward token efficiency.
Astra is good at making 3D assets, but realistically its coding was not as good as GPT 5.6 Sol.
Have we reached a point where getting better at one thing means starting to get worse at another?
This announcement is the worst one yet.
Nerfing existing usage A more expensive top-tier plan Showing off how slow Dots is right in the demo video And GPT Space—are they telling us to use it instead of Notion or what...
Overall, throughout the entire announcement, everything OpenAI wants me to look forward to just leaves me, as a developer, disappointed.
Yep. It's hitting limits because models have effectively reached the logical endpoint of how "precise" their outputs can become. In order to overcome this they need to either a) get excellent at outputting and working with massive codebases (10+ million loc) since a singificant % of systems simply cannot be done in less than that (the language itself isn't expressive enough) or b) reach the next stage of intelligence where the models are somehow able to output sequences that can solve vast complex problems with a very narrow output. Option b isn't really possible (various statistical/computational fundamental limitations forbid it) and option A is _insanely_ expensive. Which is why all of the demos the frontier labs have been pushing out are either really impressive one shot demos where the model is able to take a concise input and produce something great on its own, or very long running internal sessions where they claim to do something vast with minimal oversight (NavierStokes, C compiler, Bun rewrite).
Does anyone see the usage limits? I don't see anything indicating how much usage is in each tier.
I checked:
* https://chatgpt.com/pricing/
* https://help.openai.com/en/articles/9793128-about-chatgpt-pr...
* https://chatgpt.com/#pricing (signed in plan upgrade page)
Pure scam pricing.