Weekly limit is draining like the old 5-hour limit

What issue are you seeing?

Since the 5-hour limit disappeared, the weekly limit is dropping at about the same speed as the old 5-hour limit did.

I am using the same normal setup as before, GPT-5.5 High. I am not using Sol or any unusual heavy mode.

Before this change, a normal work session mainly reduced the 5-hour limit and the weekly limit lasted much longer. Now the weekly allowance drops very quickly during the same kind of work.

What steps can reproduce the bug?

Open Codex in VS Code.
Select GPT-5.5 High.
Use it normally on an existing project.
Check the weekly usage before and after the session.

The weekly percentage drops at roughly the rate the old 5-hour limit used to drop.

The same weekly usage value is also visible in ChatGPT Settings, so this does not seem to be only an extension display bug.

What is the expected behavior?

Removing the 5-hour limit should not reduce the total weekly capacity.

The weekly limit should still last roughly as long as it did before, instead of behaving like a single old 5-hour allowance.

Additional information

I first noticed this after the 5-hour limit disappeared from the usage page in July 2026.

My model choice and normal workflow did not change. This happens with GPT-5.5 High, so Sol is not required to see the problem.

Please check whether the weekly allowance was reduced or if the usage accounting is wrong.

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

Powered by Codex Action

You might have too many skills or MCPs. Delete those "programming-language-skills" and any obvious thing that GPT already had in it's inference. Most skills in marketplaces are not useful for frontier models, but might be useful for models with less deep reasoning.

Or your computer might be compromised, there are report of malware that steals AI token usage by proxing botnet requests through your subscription

The previous two comments are unrelated to the reported issue and are unsupported speculation. I noticed the weekly limit started draining at roughly the same rate as the former 5-hour limit when the 5-hour limit disappeared. Please keep the discussion focused on the reported usage-accounting problem and verifiable evidence.

I commented that because I cannot reproduce this bug here.

I am experiencing the same issue in the Codex desktop app.

Account and environment

Subscription: ChatGPT Pro, USD 200/month
Date observed: July 16, 2026
Client: Codex desktop app
Models used: GPT-5.6 Sol, followed by GPT-5.6 Terra in High reasoning mode

What happened

My weekly usage allowance reset on the morning of July 16. After the reset, I used GPT-5.6 Sol for a task that ran for approximately two hours. When I returned in the evening, I found that only 18% of my weekly allowance remained.

This was extremely unexpected. Under my normal usage pattern, a comparable two-hour task would usually consume less than 5% of the weekly allowance, and often no more than about 2%.

I took screenshots of the usage level and then opened a short discussion using GPT-5.6 Terra in High reasoning mode to investigate the issue. During that discussion, the remaining allowance continued to decrease by roughly 1 percentage point after every one or two messages.

The session was not unusually large or long. The total context was only about 45,000 tokens, and the model spent roughly 30 seconds reasoning. There had only been a brief discussion of a folder before this. Therefore, the rate of depletion is difficult to explain as a normal consequence of a long context window or an unusually expensive task.

After only a few additional exchanges, my remaining weekly allowance fell from 18% to 10%.

Expected behavior

The weekly allowance should decrease at approximately the same rate as it did before this change. A two-hour task and several short follow-up messages should not consume around 90% of a USD 200/month Pro account's weekly allowance.

This appears consistent with the report in this issue: after the five-hour limit disappeared, the weekly allowance began draining at a rate comparable to the former five-hour allowance. Please investigate whether the weekly quota size or the usage-accounting logic changed on July 16, or whether usage is being counted incorrectly.

After the short GPT-5.6 Terra High conversations described above, the Codex usage page still showed 10% remaining. I left the page open for more than ten minutes and the value remained unchanged at 10%. There were no visible active or queued Codex tasks.

I then closed my laptop and travelled home. During this period, I did not use Codex, ChatGPT Work, ChatGPT for Excel, or any Workspace Agent. I only used a normal ChatGPT conversation, in which ChatGPT made one GitHub connector action to post a comment to this issue.

When I later reopened Codex, the remaining allowance had fallen from 10% to 0%, and the application displayed “Your Codex and Work usage allowance is exhausted,” with the next reset shown as July 23 at 05:16.

A normal ChatGPT conversation and a GitHub connector action should not, based on the published documentation, consume the Codex/Work shared agentic allowance.

Agreed wholeheartedly. If this is a transparency regression, then add back in the 5-hour timers (👀 Tibo).

Nobody likes being peed on their leg and then being told it's actually Apple Cider when it's really not.

All the "oops I did it again" goodwill reset vibes should mathematically square with the most recent capacity equivalents of at least EIGHT full 5-hour windows per week when translated for the suspended sub-timer.

Token-use acceleration, hidden behind false largesse, will be the fastest way to lose Pro customers (👀 Tibo).

Update: this issue is still ongoing. Since the 5-hour limit disappeared in July 2026, my weekly allowance has continued to drain at an unrealistically fast rate with my usual GPT-5.5 High workflow. My model, settings and normal usage pattern have not changed.

If it helps anyone quantify what's happening: the information is already on your machine.

The rollout files under ~/.codex/sessions include used_percent for each quota window (5-hour, weekly, monthly), along with their reset times. I surface those in unsnooze usage, which renders each window as a progress bar with its exact reset date, so you can actually watch the weekly quota drain instead of only discovering you've hit the limit.

It's been much nicer than guessing how much budget is left.

https://github.com/saaranshM/unsnooze

Update: this issue is still ongoing. Since the 5-hour limit disappeared in July 2026, my weekly allowance has continued to drain at an unrealistically fast rate with my usual GPT-5.5 High workflow. My model, settings and normal usage pattern have not changed.

I agree with this and this is very obvious even without clear numbers. It is such a massive difference to before that you merely breathe in and lost 10% already, whereas when I resubbed July 7th, I had been working normally on a 150k line code project for easily 14-18 hours straight and I did not even go below 90%.

I would not say the new and current state it is roughly the 5 Hour limit but maybe like 50% slower than that. But maybe I am mistaken, didn't check yet since I was too busy.

Same workload, etc. as when I first resubbed to Codex July 7th or so, but weekly limit by now drains much faster with the same workload.

However I am getting random resets in the middle of the "week" and my reset date keeps shifting backward, so I am not that bothered yet. Even though this particular time I am quite literally having my weekly limit *exhausted 1 day into the "week" given the date kept shifting with the free irregular resets.

Now given the frequency of those, I assume I will get another random reset for no apparent reason before that actual weekly reset date.

But we will see. If not, then this would obviously be horrible given nothing in terms of the workload changed.

Turns out it increased a little bit on average, since I had two workflows running for two of those days but compare this with early days of july, as I said, I barely got even below 90% weekly usage in the first 2-3 days combined.

Then after those "free resets" in the middle of the week, 2 days ago or so I ended up using 100% of my weekly allowance in just one day. lol

To me that is not explicable by this graph.

Title: Weekly usage quota depleting significantly faster since the 5-hour limit was removed (July 2026)

Description:
Since the temporary removal of the rolling 5-hour usage limit on July 12, 2026, my weekly usage quota is draining at a much higher rate than before, despite no change in my usage pattern.

What I observed:

The daily/5-hour usage indicator has disappeared from the usage page.
The weekly quota is depleting noticeably faster than it did when the 5-hour and weekly limits operated separately.
I am paying the same subscription price but effectively receiving less usable capacity.
My workflow and model selection have not changed.

Expected behavior:
Removing the 5-hour limit should not reduce total weekly capacity. The weekly allowance should last roughly as long as it did before, not behave like a single old 5-hour bucket stretched across a week.

This appears to match a known issue already reported by other users (see openai/codex issue #33685), where the same behavior was observed independent of using Sol or any unusual workload.

Additional update after the latest Codex app update

I observed another unexplained quota reset / reset-date change today.

Before installing the latest Codex app update, my usage meter had continued to decrease during normal work:

earlier state: 78% remaining, reset July 26, 2026;
immediately before the update: 45% remaining, reset July 26, 2026.

After downloading and installing the latest Codex app update, the meter changed to:

100% remaining, reset July 28, 2026.

This was not the expected July 26 reset. Both the remaining percentage and the reset date changed at the same time. I cannot determine from the UI whether this was:

an intentional compensation/reset;
a server-side quota recalculation;
a correction to usage accounting;
or merely a client refresh that exposed a server-side change.

It would be helpful if OpenAI clarified whether users affected by this issue are receiving automatic quota resets and whether the weekly accounting logic was changed in the latest rollout.

I also have one model-specific observation. For the bounded tasks I tried, GPT-5.6 Luna appeared to produce results that were not materially worse for those tasks while consuming very little visible weekly allowance. This was not a controlled benchmark, so I am not claiming general model equivalence. However, the contrast was noticeable: Luna usage barely moved the meter, while GPT-5.5 Very High and GPT-5.6 Terra/Sol had previously reduced it very quickly.

This reinforces the need for per-session usage attribution by model and reasoning level. Without it, users cannot tell whether these large changes are expected model pricing, retries/compaction, a quota reset, or a metering regression.

@AlexMyln

@AlexMyln

I noticed the same today. My remaining allowance also reset to 100%.

Banked reset tokens become a guessing game without separate window visibility

I am a ChatGPT Plus user and I want to add a related but distinct UX issue that has cost me real money.

I had 3 banked "Full reset" tokens saved for when I truly needed them. After the 5-hour meter disappeared from the UI, I kept seeing the combined gauge drop to 50‑60%, then jump back to 100% a few hours later. Since the gauge is labelled solely as "Weekly", I interpreted these jumps as some kind of partial weekly refresh.

I kept working. The gauge dropped again. I burned a banked reset. Again the same pattern — drop to ~50%, then back to 100%. I burned a second banked reset. Only later — after reading the account/rateLimits/read response shape (#24080) — did I realise those "jumps" were the 5-hour window auto‑resetting, not a weekly refresh. The UI simply never told me.

The result:

2 of my 3 banked "Full reset" tokens were used not because my weekly quota was truly exhausted, but because the combined gauge gave me no way to distinguish which window was depleted.
My 3rd and last token expires August 12. I am now at 0% and cannot tell if I should burn it or wait for the 5-hour auto‑reset, because the UI still shows only one percentage.

This is not just a "weekly drains too fast" accounting issue. It is a transparency issue:

Banked resets are scarce and have expiration dates — treating them as a "click to unblock" button is fine, but treating them as a discovery tool for hidden quota windows is not.
A user should be able to see each active quota window separately (5-hour, weekly) with its own percentage and reset time, especially when they own banked resets that affect both windows differently.
If the 5-hour window is truly merged into weekly accounting, then the banked resets should not claim to "reset all limits" when the UI cannot confirm what happened.

I understand from #24080 that account/rateLimits/read returns primary (300 min) and secondary (10080 min) with separate usedPercent and resetsAt. That data already exists on the backend. Exposing it in the UI — or at least in the status bar — would prevent users like me from wasting real banked resets on phantom quota problems.

ChatGPT Plus, Windows, Codex Desktop + VS Code extension. Observed July 12–23, 2026.

@IvoDimitrovPro

@IvoDimitrovPro

I also wasted one reset before the 5-hour meter disappeared. I thought it would only reset the 5-hour limit, but then I saw that it also reset the weekly limit.

So my case was slightly different, but I agree that the reset behavior and the separate quota windows should be explained much more clearly.

@IvoDimitrovPro

@IvoDimitrovPro I dont think what we are seeing here is a misslabled 5h window, at least not for me. It does reset randomly from time to time but I am at 0 now since 12h. It drained me a full reset within 4 hours with light usage and only using 5.5. Anyway... will now go back to Claude because this is a real deal breaker atm.

@IvoDimitrovPro I dont think what we are seeing here is a misslabled 5h window, at least not for me. It does reset randomly from time to time but I am at 0 now since 12h. It drained me a full reset within 4 hours with light usage and only using 5.5. Anyway... will now go back to Claude because this is a real deal breaker atm.

I am already back in claude and Claude also has some shady practices going on. For example on Opus 5 they have a feature flag they evaluate which if false includes a so called "Opus 5 Prompt bundle" including 2 sentences that explicitly prevent Opus 5 from spawning Dynamic Workflows / Subagents even when you explicitly chose the ultracode mode, while burning more usage at ultracode as normal.

They also explicitly set some env to "no_nudges" so either it or the user is never nudged that subagents / dynamic workflows whatever the right term is exist.

I noticed this after some session yesterday.

Those sentences in the system prompt explicitly say to never do it unless the user explicitly requests it. And Ultracode does not count as the user requesting it.

My weekly quota got burned in just 2 hours previously I could run for days same models same project same cycle. So something is wrong on OpenAI side.

Some recent reports outside GitHub show that people are still seeing the same or very similar problem:

GPT Pro 5x stopped feeling like a 5x plan after the five-hour limit was removed - same workload used to last much longer, and the current weekly allowance feels close to the old 5-hour allowance. GPT-5.5 is also specifically mentioned.
Codex removed the 5-hour limit... and gave it a weekly nametag - users in the comments also report that this is not limited to the new models, including the same fast depletion with GPT-5.5.
Codex Usage Limits and Performance Megathread - multiple users report much faster weekly depletion than before, in some cases losing a large part of the weekly allowance in only a few hours.

I am not posting these as proof of the exact cause, just to show that this problem is not isolated to this GitHub issue and people are still reporting it now.

I think at this point this deserves an official response and investigation.

I am experiencing the same issue on a ChatGPT Plus account, and I can reproduce it across every Codex client connected to my account.

Account and environment

Subscription: ChatGPT Plus
Platform: Windows
Codex CLI version: 0.146.0
Model during the affected session: GPT-5.6 Terra, reasoning low
Clients checked:
- Codex VS Code extension
- Codex desktop application
- Codex CLI
- ChatGPT/Codex web usage page
Date observed: August 4, 2026

What happened

During a working session of approximately three hours, my weekly Codex allowance dropped to 19% remaining, meaning approximately 81% of the weekly allowance was consumed.

At no point did Codex display a five-hour allowance, warn that I was approaching a five-hour limit, or prevent me from starting additional tasks because a five-hour window had been exhausted.

The weekly allowance appears to be draining directly and at a rate comparable to what would previously have been expected from a shorter five-hour usage bucket.

Missing five-hour bucket

The five-hour usage meter is absent from every interface I checked.

Running /status in Codex CLI 0.146.0 displays only:

Weekly limit: 19% left (resets 22:59 on 8 Aug)

It does not display:

A five-hour or 300-minute usage bucket
The percentage remaining in that bucket
Its reset time
Any indication that a five-hour limit is being enforced

The same weekly-only information appears in the VS Code extension, desktop app and web usage page.

`/usage daily` does not explain the depletion

I also ran:

/usage daily

It displayed historical token activity:

Lifetime 783M
Peak 222M
Streak 9d
Longest task 12h 22m

However, it does not show:

How many tokens or usage units were charged during the affected three-hour session
How those tokens translated into approximately 81% of the weekly allowance
Whether a 300-minute bucket is provisioned on my account
Whether a model or tool multiplier was applied

Expected behavior

A normal three-hour working session should not consume approximately 81% of the entire weekly Plus allowance without any visible five-hour-limit warning or interruption.

If the five-hour bucket has been intentionally removed, the weekly allowance should not behave as though it has replaced a single former five-hour allowance.

Requested investigation

Please investigate:

Whether the weekly allowance size or accounting logic changed when the five-hour meter disappeared.
Whether a 300-minute usage bucket is currently provisioned and enforced on Plus accounts.
Why the five-hour bucket is absent from the CLI, desktop app, VS Code extension and web usage page.
Whether my affected session was charged using an unexpected multiplier.
Whether the account usage ledger can be audited for abnormal or duplicated deductions.

I have also submitted this issue to OpenAI Support and can provide sanitized screenshots showing the weekly-only /status output and the missing five-hour meter.

Yes, i'm also facing the same, for a small task it used 20% of weekly limits,
The only reason we're using Codex over Claude was that it was little generous, now there's no reason to use Codex.

I just ran into the same issue. On a plus account with Terra xHigh, it burned whole weekly usage in one session. With 30% in one prompt. Eats tokens faster than Fable 5 on ultracode. This was my first time trying out Codex, because I wanted Claude alternatives, but I guess I'm headed back to Claude Code.

Update after my latest weekly reset:

After about one day of moderate use, 51% of my weekly limit is already gone.

I used mostly Terra Medium, and less Sol High than before. This was not a heavy usage day. At this rate, the weekly limit would not even last two normal working days for me.

Also, in Settings > Usage, where I previously had reset options, I now have no reset available at all. The only extra option shown there is to buy credits.

There are many similar reports from just the last week:

GitHub - weekly usage / quota problems:
#40067
#39167

Reset availability:
#39760
https://www.reddit.com/r/codex/comments/1vkfcnz/open_ai_removes_banked_usage_resets_from_codex/

This does not look like normal weekly usage anymore.

+1, I'll write about this on Reddit — I suspect it'll get a lot of attention.

There's no "Exec" entry at all in my ccusage logs... so this looks like a silent billing skip rather than a visible one.

Either way, my limits ended up completely drained.

This looks like the same root cause reported in several other open issues — possible duplicates or close variants of this one:

Codex Chronicle silently drains subscription quota 24/7 and triggers auth token invalidation #27773 — Chronicle silently drains subscription quota 24/7
Chronicle stays enabled after a Pro → Plus downgrade, consuming the smaller allowance in the background #36805 — Chronicle stays enabled after a plan downgrade, draining the smaller allowance
Chronicle never generates 6-hour summaries — only the 10-minute tier ever runs #37983 — Chronicle never generates 6-hour summaries, only the 10-minute tier runs
[macOS][Computer History] Recorder creates Skysight segments but 10-minute summaries never start #40386 — Recorder creates segments but 10-minute summaries never start
Weekly limit is draining like the old 5-hour limit #33685 — Weekly limit draining the way the old 5-hour limit used to

And related (same "silent background drain" pattern, not necessarily Chronicle specifically):

Opening the ChatGPT desktop app silently consumes the Codex weekly limit (fixed 6% per background suggestion run; reproduced under controlled conditions) #37445, [Privacy] Codex initiates screen capture while no task is running, offline, and usage quota is exhausted #35754, Desktop-bundled Codex: guardian_subagent approvals reviewer enabled without consent; background memory jobs drain plan quota while idle #36733, Paid x5 plan loses ~20% of 5-hour limit while idle #31796, Codex usage quota decreases slowly even when I am not actively using Codex #26600, Plus: quota hits 100% after idle period with zero sessions; Jul 2 single-prompt drain 1%→100% in 6 min (session logs) #31174

Important update:

The 5-hour limit was restored for Plus users on August 25.

This directly affects the original issue reported here, because the problem started after the 5-hour limit disappeared and the weekly limit began draining roughly like the old 5-hour limit.

Now that the 5-hour limit is back, I will watch the weekly usage closely to see whether the weekly limit also starts draining more slowly again, or whether the same abnormal consumption continues.

I will update this issue again when I have enough usage to compare it properly.

Important update:

The 5-hour limit was restored for Plus users on August 25.

This directly affects the original issue reported here, because the problem started after the 5-hour limit disappeared and the weekly limit began draining roughly like the old 5-hour limit.

Now that the 5-hour limit is back, I will watch the weekly usage closely to see whether the weekly limit also starts draining more slowly again, or whether the same abnormal consumption continues.

I will update this issue again when I have enough usage to compare it properly.

Finally! This is our common victory! Brothers!

Navigation Menu

Repository navigation

Weekly limit is draining like the old 5-hour limit #33685

What issue are you seeing?

What steps can reproduce the bug?

What is the expected behavior?

Additional information

Account and environment

What happened

Expected behavior

4 remaining items

Additional update after the latest Codex app update

Banked reset tokens become a guessing game without separate window visibility

Account and environment

What happened

Missing five-hour bucket

`/usage daily` does not explain the depletion

Expected behavior

Requested investigation

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Weekly limit is draining like the old 5-hour limit #33685

Description

What issue are you seeing?

What steps can reproduce the bug?

What is the expected behavior?

Additional information

Activity

github-actions commented on Jul 16, 2026

3esmit commented on Jul 16, 2026

3esmit commented on Jul 16, 2026

mikk73 commented on Jul 16, 2026

3esmit commented on Jul 16, 2026

CarolineCheng233 commented on Jul 16, 2026

Account and environment

What happened

Expected behavior

naive4E4A55 commented on Jul 16, 2026

jasonzorn commented on Jul 18, 2026

mikk73 commented on Jul 18, 2026

saaranshM commented on Jul 19, 2026

tobiasbarthold commented on Jul 19, 2026

Orky-team commented on Jul 20, 2026

4 remaining items

AlexMyln commented on Jul 21, 2026

Additional update after the latest Codex app update

mikk73 commented on Jul 21, 2026

IvoDimitrovPro commented on Jul 23, 2026

Banked reset tokens become a guessing game without separate window visibility

mikk73 commented on Jul 23, 2026

c-schober commented on Jul 23, 2026

tobiasbarthold commented on Jul 26, 2026

phyrexia commented on Jul 30, 2026

mikk73 commented on Jul 30, 2026

juancamilotorres111-commits commented on Aug 4, 2026

Account and environment

What happened

Missing five-hour bucket

/usage daily does not explain the depletion

Expected behavior

Requested investigation

koderkashif commented on Aug 11, 2026

jrbarhydt commented on Aug 17, 2026

mikk73 commented on Aug 22, 2026

axisrow commented on Aug 30, 2026

mikk73 commented on Aug 30, 2026

axisrow commented on Aug 31, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

`/usage daily` does not explain the depletion