codex - 💡(How to fix) Fix Usage not decreasing at all [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openai/codex#17455Fetched 2026-04-12 13:28:31
View on GitHub
Comments
1
Participants
2
Timeline
7
Reactions
0
Author
Timeline (top)
labeled ×4unlabeled ×2commented ×1

Root Cause

I might be missing something, but I've checked multiple times and believe something may be wrong on the backend side. Codex definitely recognizes usage on the account, because it starts counting down the 5h reset window (e.g., "4h 36m" instead of "5h"). But the 5h limit stays at 100% (and the weekly limit remains unchanged) even though similar usage was very noticeable yesterday.

RAW_BUFFERClick to expand / collapse

I've been running ~25 parallel Codex CLI instances with exec for about two hours, and neither my 5h nor weekly limits have decreased even by 1%. Beyond that, the "Usage breakdown" in Codex Web doesn't show any activity for today (april 11) at all.

I'm not entirely sure what led to this. I've been using the base_instructions and developer_instructions config parameters, which I wasn't doing before. It could also be something else that isn't clear to me.

I might be missing something, but I've checked multiple times and believe something may be wrong on the backend side. Codex definitely recognizes usage on the account, because it starts counting down the 5h reset window (e.g., "4h 36m" instead of "5h"). But the 5h limit stays at 100% (and the weekly limit remains unchanged) even though similar usage was very noticeable yesterday.

I've tested this on two (pro) accounts, and both stay at 100% after a significant number of tokens used this way: user-yd2KvMYwDwnAw1P5XaRaJTD1 and user-dZfuXwn5gK9c3D8UUV4oV6B1.

Happy to provide any additional information or run further tests if that would help narrow down the issue.

extent analysis

TL;DR

The issue may be related to incorrect usage tracking or reporting on the Codex backend side, and further investigation is needed to determine the root cause.

Guidance

  • Verify that the exec command is correctly configured and that the base_instructions and developer_instructions config parameters are being used as intended.
  • Check the Codex documentation to ensure that the usage limits and tracking are working as expected, especially with parallel CLI instances.
  • Test the usage tracking with a single CLI instance to see if the issue persists, and compare the results with the parallel instances.
  • Provide additional information about the exec command and the config parameters used, as well as the exact commands run, to help narrow down the issue.

Notes

The issue may be specific to the exec command or the config parameters used, and further testing and investigation are needed to determine the root cause.

Recommendation

Apply workaround: Run tests with a single CLI instance and monitor usage tracking to see if the issue persists, and report the results to help narrow down the issue.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING