Codex 28 · Theme · #40

Model-dependent usage limits

10people describe this

10 on X

1 win · 5 problems · 1 idea · 3 mixed · first reported 15 Oct 2026

X replies to the promise post are counted apart from Rantr votes. “This happened to me too” adds your browser to this theme on Rantr, once.

Where OpenAI stands

Not addressed yet

No recorded announcement has answered this yet.

In their words

What people said

Problem
With 5 hours limit: Claude pro can make three videos with opus 5.5 in level Extra high, Codex plus cannot finish even one videos with 6.1 sol extra high. I have tried it many times. OpenAI gives reset to compensate this weakness. Can OpenAI gives reset all the time?

@Looking44truth on X ↗ · 26 Oct 2026
Problem
Something is wrong with gpt models, Opus 5.5 can gives you the room to work while gpt models (except for luna) burns 5h limits in like 10-20min if lucky and wouldn't even do the task completely🤣 how the fck Opus 5.5 is 10x cheaper to consume and 10x better in quality.

@hellobyedot on X ↗ · 25 Oct 2026
Mixed
Stupid idea cuz i can use opus for a solid hour, doing better work than astra which hits 5hr limit in like 20 mins.

@jessehill12333 on X ↗ · 21 Oct 2026
Problem
I just used up my 5 hour limit in an hour just to consolidate two pdfs. I have a plus subscription and used 6.1 Sol med. Decided to test the same exact job with Sonnet 5.5 med and I still have 60% of usage after 20mins with the same output. Sad given I prefer oAI 😔

@thoughtfulgubi on X ↗ · 20 Oct 2026
Mixed
I gave gpt 6.1 sol a few simple builds that took it over a week and all the resets to for it to not even get close to finishing them.. I give opus 5.5 the same build, almost completely done and used 5% of my usage.. @thsottiaux I ran into lots of errors and more.

@LewBdf99567 on X ↗ · 19 Oct 2026
Mixed
I'm a Claude Max and GPT Pro user, but I've noticed a significant difference in usage for identical tasks. I used Opus 5.5 Max effort for continuous development without reaching the 5-hour limit, while Astra Extra High effort easily burns through nearly 1/3 of my weekly usage

@ffzh37119 on X ↗ · 19 Oct 2026
Problem
You guys know perfectly that your service doesn't keep up with the competence. Opus 5.5 20$ limits are absurd, while 6.1 sol burns the week in 2 days of normal usage. You're just behind.

@Arzuparreta on X ↗ · 17 Oct 2026
Idea
It’s laughable, move to Claude max and you get double the usage of ChatGPT 500 plan. Run Opus 5.5 on high instead of Sol 6.1 Max (it’s smarter), it lasts all week and I still have 60% left and costs less than half the price.

@wolfswut777 on X ↗ · 17 Oct 2026
Win
I switch to claude $200 plan today and difference is shocking 😧 its 10 times faster and almost 10x more usage then codex $200 plan.

@burhanwork on X ↗ · 16 Oct 2026
Problem
Why does GPT have so much less quota than Claude? I don't get it. I remember it was the other way around half a year ago. At that time, GPT was really durable - for the same task, Claude could go 2-4 hours while GPT would use up its five-hour quota in just one hour

@LoooUmmi14040 on X ↗ · 15 Oct 2026

Your experience

Had this too? Say what happened.

JEV reads your words as you type and shows which theme they belong to. It may place them in another theme if that fits better.

What happened with “Model-dependent usage limits”?

0 / 1200

Tell us a little more — at least 20 characters.

Your words stay yours.

AI reads your draft as you type. Only what you publish is public. Privacy

Not sure where to start?

Think of the last task you tried. What were you doing, what happened, and what would have helped? You can also say you haven't tried a change yet.

01
What you said

Read as you type: a win, a problem or an idea, and the words that carry it.
Ready when you are
Who else had it

The theme your words join, and how many people on Rantr and X described it.
Where OpenAI stands

Shipped, reset, or not addressed yet, from the recorded announcements.
What to do now

A step the announcements make possible for you, when there is one.

Status

Where OpenAI stands

“Shipped” means an announcement was linked to the theme by AI, not that the problem is fixed for everyone. “Not addressed yet” means no recorded announcement was linked; it is not proof that nothing changed.

No recorded announcement has been linked to this theme yet.

Corrections

Changes to this theme

20 Oct 2026, 02:34 UTC

Renamed from “Allowance lasts differently by model” to “Model-dependent usage limits” as more people described it.
17 Oct 2026, 00:36 UTC

“Quota balance shifts between models” was merged into this theme: JEV judged them the same problem.

Follow this theme by email

We send a confirmation link first; nothing else is sent until you confirm. Then at most one email a day, only when the theme changes.

X thread: 2769 replies to the promise post collected (4 Oct 2026 – 26 Oct 2026): 2555 placed in a theme and 214 set aside as off-topic or unsafe. X replies are found by search and each one is checked against its public post on X. Last read: 26 Oct 2026, 23:46 UTC. Not every reply is read.

← Every theme