← sources

So are we just going to forget about this shared ChatGPT/Codex usage nonsense? (NO!)

source: reddit (reddit:ChatGPT)published 2026-11-09 18:06ingested 2026-11-10 02:37

retained text

So are we just going to forget about this shared ChatGPT/Codex usage nonsense?

Haven't seen much discussion about this lately, and December is coming up pretty damn fast, so figured I'd bring it up again.

For anyone who missed it, OpenAI is planning to combine ChatGPT's advanced reasoning usage and Codex/Work usage into one shared allowance.

Right now they're seperate. I can run Codex right into the ground (and regularly do lol) and still go back to ChatGPT and use High reasoning for research, troubleshooting, working through complicated problems, whatever.

Under the new system, both are going to eat from the same pool.

And here's the fun part. We STILL don't know how big that pool is, or how much different models are going to consume from it.

OpenAI says most people won't hit limits any more frequently than they do now.

Ok? Maybe that's true. I'd love it if it were. But how the hell are we supposed to know when they won't tell us what we're actually getting?

December is a few weeks away. We're apparently just supposed to trust them, and find out when the change happens.

And to the inevitable "just pay more then" crowd, yes, I'm aware that computing resources cost money. I'm aware OpenAI is a business. I've been paying that business for years, believe it or not.

The point isn't that I expect unlimited access to everything for $20.

The point is that I'm already paying for a service that works for me, and they're changing how much of that service I can use without telling me what I'm actually going to get.

If maintaining my current usage ends up requiring more money, that's a price increase for my use case, however they want to dress it up. And I think I'm entitled to be annoyed about that.

And "you can always upgrade" isn't really an answer to whether an existing subscription is getting worse value. Of course I can pay more. I can also pay somebody else, which is increasingly looking like a pretty sensible option.

I like ChatGPT. I use it constantly. Which is why I'm pissed off about the possibility of losing access to something I've been paying for and enjoying for years.

I don't care if they want to integrate Chat, Codex and Work into one big happy family. Go nuts. I just don't see why that means the usage allowances have to be combined too.

Maybe they'll give us enough that it won't matter. But given the way these companies operate, you'll forgive me if I'm not overflowing with confidence here. With an IPO apparently on the horizon and all the pressure to improve their numbers, I'm not exactly overflowing with confidence that customers are going to come out ahead here.

And I'm certainly not interested in paying more just to get back something I already had.

If everything's going to be fine, publish the numbers! Its not complicated.

And if it's going to mean less usage for the same money, at least have the balls to tell us before it happens.

I don't think we should just let this disappear from the conversation because the initial outrage died down. This could have a real impact on people who use both products extensively, and we're getting awfully close to finding out the hard way.

I've enjoyed ChatGPT for years. I'd quite like to keep enjoying it, without having to wonder what part of my subscription is going to get worse next. I don't particularly want to wake up one morning in December and discover that coding for a few hours means I can't have a proper conversation with ChatGPT until some timer resets.

Official announcement:

https://help.openai.com/en/articles/20001625-understanding-chatgpt-and-codex-usage-limits

document fetched 2026-11-10 00:20 · canonical https://reddit.com/r/ChatGPT/comments/1xryxa8/so_are_we_just_going_to_forget_about_this_shared