5h limits totally broken

Open 💬 4 comments Opened Jul 7, 2026 by Howchie
💡 Likely answer: A maintainer (github-actions[bot], contributor) responded on this thread — see the highlighted reply below.

I've hit my 5h limits multiple mornings now on very light work using 5.4. I'm on the plus plan and usually never hit 5h limits. So, adding my voice, usage is totally messed up. The circled days are the two I have hit the 5hr cap and it is clearly nowhere near the usage of previous days (I only work 9-5, but it's very obvious even stacking 5x perfectly complete usage of every 5hr cap couldn't reach some of the other days usage).

<img width="1831" height="491" alt="Image" src="https://github.com/user-attachments/assets/c1aaa2b5-e499-4ade-b622-d434a7d1462f" />

It's also not because I'm using a high model as these days I was using lighter models predominantly because the tasks I ran were straightforward and not using big contexts.
<img width="1855" height="627" alt="Image" src="https://github.com/user-attachments/assets/ec4c38da-ba96-4b0f-ad26-27a23ca99ac1" />

View original on GitHub ↗

4 Comments

github-actions[bot] contributor · 1 month ago

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

  • #31199
  • #30918
  • #31174
  • #30694
  • #30939

Powered by Codex Action

astral3733 · 1 month ago

Me, too; and it's getting worse. This morning, Codex had an update. After that, I wasted 9-10% of my five-hour usage with 2 minutes for one simple bug fix. I have realized that over the past 1-2 weeks, every time OpenAI had an update, the problem got worse. I am also a Plus user. BTW, I only use "Low" for all tasks, and I have never used agents.

faction-hash · 1 month ago

I think this might actually be a bug, because the context seems to get completely wiped out after just two messages. However, when I use the 5.4 Mini model, it lasts significantly longer—which makes me wonder if the behavior is intentionally engineered rather than a glitch.

To investigate further, I complained about this issue about a week ago and was given extra tokens, but those were also depleted quickly. I also tested it on a brand-new account and exhausted its entire credit limit in 2 chat, which burnt only 3 million tokens. In contrast, my older account has a total limit of 7.2 billion tokens burnt in 2 chat. Given such a massive discrepancy between accounts, I strongly suspect this is a bug.

Howchie · 1 month ago

Yeah I'm not sure how to check exact tokens but today was just a planning
job, as in refactor an md file with notes into a plan, used 5.4 medium and
that burned my 5hr. It's clearly not the usage I've experienced on other
days, my two lowest usage days in weeks are the ones that "hit the limit".
Hopefully a bug and a reset soon.

On Tue, 7 July 2026, 5:21 pm faction-hash, @.***> wrote:

faction-hash left a comment (openai/codex#31345) <https://github.com/openai/codex/issues/31345#issuecomment-4902208734> I think this might actually be a bug, because the context seems to get completely wiped out after just two messages. However, when I use the 5.4 Mini model, it lasts significantly longer—which makes me wonder if the behavior is intentionally engineered rather than a glitch. To investigate further, I complained about this issue about a week ago and was given extra tokens, but those were also depleted quickly. I also tested it on a brand-new account and exhausted its entire credit limit in 2 chat, which burnt only 3 million tokens. In contrast, my older account has a total limit of 7.2 billion tokens burnt in 2 chat. Given such a massive discrepancy between accounts, I strongly suspect this is a bug. — Reply to this email directly, view it on GitHub <https://github.com/openai/codex/issues/31345?email_source=notifications&email_token=AC5SRY23J7JVEKUQ2BODLZL5DS6H7A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOJQGIZDAOBXGM2KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L#issuecomment-4902208734>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC5SRY7VUUAU7AHRFXFNKP35DS6H7AVCNFSNUABFKJSXA33TNF2G64TZHM4TMNJUGE2TMNBZHNEXG43VMU5TIOBSGQ4DONBSGQ32C5QC> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/AC5SRY3JG5NOJUEWQO3EUMD5DS6H7A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOJQGIZDAOBXGM2KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KUZTPN52GK4S7NFXXG> and Android <https://github.com/notifications/mobile/android/AC5SRY3ZAP67QACLFUJD25L5DS6H7A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOJQGIZDAOBXGM2KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2K4ZTPN52GK4S7MFXGI4TPNFSA>. Download it today! You are receiving this because you authored the thread.Message ID: @.***>