# Codex Pro 5x weekly limit exhausted in 2 days with unchanged workflow + purchased credits not applied

**URL:** <https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726>\
**Category:** Support\
**Tags:** codex\
**Created:** [August 16, 2026, 2:14pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726 "2026-08-16T14:14:52Z")\
**Posts on this page:** 8\
**Page:** 1

<div class="post-metadata">

**Author:** ![kharitonova.pro](https://avatars.discourse-cdn.com/v4/letter/k/57b2e6/32.png) [@kharitonova.pro](https://community.openai.com/u/kharitonova.pro)\
**Post date:** [August 16, 2026, 2:14pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/1 "2026-08-16T14:14:52Z")

</div>

I would like to report a significant change in Codex usage behavior.

I have been using Codex continuously since April/May 2026. My workflow and project complexity have been consistent: I use Codex for developing my application and platform, analysis, iterations, and improvements.

In May, I was able to work extensively even on the $20 plan. After upgrading to the $100 plan, during June and July I worked daily without being blocked by limits.

Since August, with the same working style and similar tasks, my weekly limit is exhausted within 2–3 days. I am not running significantly more tasks than before, but the effective capacity has dropped dramatically.

Additionally, I purchased $20 Codex credits. The payment succeeded, but after my included limit was exhausted, the purchased credits were not applied automatically. The dashboard shows 0 credits and no usage record explaining where they went.

Please investigate whether there is a usage metering, quota assignment, or credit activation issue affecting my account

---

<div class="post-metadata">

**Author:** ![trojancode](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/trojancode/32/757526_2.png) [@trojancode](https://community.openai.com/u/trojancode)\
**Post date:** [August 17, 2026, 3:33pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/2 "2026-08-17T15:33:41Z")

</div>

I’m facing the same issue. Previously, I was on the **$20 plan** , and I was able to complete my daily work without hitting the usage limits.

Then I upgraded to the **$100 plan**. Initially, throughout July, it was very helpful—I wasn’t even reaching 50% of my usage limit. But now, with the same workflow and similar tasks, I’m hitting the usage limit much faster. It almost feels like I’m back on the $20 plan.

I’ve tried using different models and different effort levels. I even switched back to **5.5**. When I reduce the effort level, the quality of the output decreases, but the usage still seems to be about the same.

So I’m not sure what changed, but the $100 plan now seems to reach its limits much faster than it did before, even though my usage pattern hasn’t really changed.

---

<div class="post-metadata">

**Author:** ![itsckl](https://avatars.discourse-cdn.com/v4/letter/i/ecae2f/32.png) [@itsckl](https://community.openai.com/u/itsckl)\
**Post date:** [August 17, 2026, 6:41pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/3 "2026-08-17T18:41:47Z")

</div>

same here. Issue started last Thursday.

---

<div class="post-metadata">

**Author:** ![PaulBellow](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/paulbellow/32/597962_2.png) [@PaulBellow](https://community.openai.com/u/PaulBellow)\
**Post date:** [August 17, 2026, 6:51pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/4 "2026-08-17T18:51:40Z")

</div>

> [@trojancode](#):
>
> _seems_ to reach its limits much faster

Not very scientific! 😅

A valuable member in our community, @merefield, has a tool that might help you pin down exactly if you’re running more tokens or not.

> [@Introducing Codexometer: keep track of remaining quota, session telemetry & model benchmarks!](https://community.openai.com/t/introducing-codexometer-keep-track-of-remaining-quota-session-telemetry-model-benchmarks/1389878/):
>
> [Codexometer](https://github.com/merefield/codexometer) is a small, retro terminal dashboard for your current Codex quota. Leave it open in a second terminal window or pane to see every active usage window, its remaining capacity, and its reset time—without repeatedly opening /status in the Codex session where you are working. Why use it? /status is useful, but it lives inside your active session. Codexometer turns the same quota information into an always-visible companion display that refreshes automatically and uses your pre…

---

<div class="post-metadata">

**Author:** ![kharitonova.pro](https://avatars.discourse-cdn.com/v4/letter/k/57b2e6/32.png) [@kharitonova.pro](https://community.openai.com/u/kharitonova.pro)\
**Post date:** [August 18, 2026, 2:09pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/5 "2026-08-18T14:09:09Z")

</div>

Yes, this is exactly what I have been seeing as well.

My workflow did not materially change, but the weekly Codex allowance started being consumed dramatically faster than before.

There is another serious issue in my case.

After my included limit was almost exhausted, I purchased 500 additional Codex credits for $20. When the included limit reached 0%, those 500 credits never became available.

OpenAI Support later claimed that the credits had been applied to a negative usage balance. I asked them to provide the actual transaction-level records showing the task, usage, negative balance, and deduction of the 500 credits.

They confirmed that they could not provide those records.

Despite that, they refused to restore the 500 credits, refused to refund the $20, and refused to escalate the case to Billing/Payments or a supervisor.

So there appear to be two related transparency problems here:

1. Included Codex limits are now being consumed much faster with essentially the same workflow.

2. Purchased credits can disappear without the user being able to see a corresponding transaction explaining exactly where they went.

My support case is #13278399.

I think OpenAI needs to investigate the metering itself, not just tell individual users that they have reached their limits.

---

<div class="post-metadata">

**Author:** ![NASTUDIO.CY](https://avatars.discourse-cdn.com/v4/letter/n/e47c2d/32.png) [@NASTUDIO.CY](https://community.openai.com/u/NASTUDIO.CY)\
**Post date:** [September 22, 2026, 9:57am UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/6 "2026-09-22T09:57:47Z")

</div>

I believe the problem started when OpenAI greyed out Pro 20x.

My impression is that, due to capacity overload, OpenAI significantly reduced usage limits across Plus, Pro 5x, and Pro 20x without clearly informing customers. Pro 5x now feels like the old Plus, and Pro 20x feels like the old Pro 5x.

What is unacceptable is the lack of transparency. Usage is disappearing much faster even with simple tasks, while support keeps replying with generic messages instead of explaining what changed.

If this is a temporary capacity issue, OpenAI should clearly say so and provide a timeline. As a paying professional user, I feel I am now receiving significantly less usage for the same subscription.

If Claude releases a model comparable to Astra for my professional workflow, I would seriously consider moving my entire office to Claude. OpenAI has been extremely disappointing in how this situation has been handled.

---

<div class="post-metadata">

**Author:** ![OpenAI\_Support](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/openai_support/32/561845_2.png) [@OpenAI\_Support](https://community.openai.com/u/OpenAI_Support)\
**Post date:** [October 17, 2026, 10:04pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/7 "2026-10-17T22:04:06Z")

</div>

Hi @[**kharitonova.pro**](https://community.openai.com/u/kharitonova.pro) - I understand the concern about faster limit consumption and purchased credits being unavailable.

You can check your balance, recent usage, and reset information in the [usage dashboard](https://chatgpt.com/codex/settings/usage). Usage varies by the model, context, reasoning effort, speed, tools, and task complexity, but that alone does not explain the change reported here. Please refer to [Codex usage guidance](https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan#usage-limits-by-plan:~:text=task%2Dspecific%20endpoints.-,Usage%20limits%20by%20plan,-If%20you%E2%80%99re%20nearing).

Running tasks can leave a negative credit balance that later purchases first cover ([details](https://help.openai.com/en/articles/12642688-using-credits-for-flexible-usage-in-chatgpt-personal-plans#:~:text=not%20automatic%20reloads.-,Managing%20credit%20balance,-If%20a%20task)). Confirming your deduction requires a private review. Please follow up in your existing support thread or contact [support@openai.com](mailto:support@openai.com) with balance/usage screenshots, dates and time zone, and the Codex client/model used. Keep billing and account details private.

~Saisha

---

<div class="post-metadata">

**Author:** ![SaishaB](https://avatars.discourse-cdn.com/v4/letter/s/d2c977/32.png) [@SaishaB](https://community.openai.com/u/SaishaB)\
**Post date:** [October 17, 2026, 10:04pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726/8 "2026-10-17T22:04:15Z")

</div>

Closed on October 17, 2026, 10:04pm UTC


