Codex Usage

Codex Pulse

What Codex users are asking right now

The most reported Codex usage problem this week is unexpected usage depletion, with 15 public GitHub reports in the last 7 days.

Scheduled observation Covers Sep 23 to Oct 7

Latest scheduled observation

Observed

199 GitHub community reports observed by Codex Pulse.

0 change in reports since the previous scheduled observation.

Usage-related issues in openai/codex, within Pulse's 14-day creation window. Counts describe this source set, not affected users or platform-wide OpenAI behavior.

Previous scheduled observation: (199 reports).

Scheduled scan record on GitHub
Recent scheduled observations
29302026-09-30 20:49:27.091 UTC: 293 reports2026-10-01 05:56:33.785 UTC: 285 reports2026-10-01 15:32:59.696 UTC: 255 reports2026-10-01 21:02:27.345 UTC: 252 reports2026-10-02 05:39:30.130 UTC: 247 reports2026-10-02 14:45:06.712 UTC: 240 reports2026-10-02 20:46:15.632 UTC: 241 reports2026-10-03 05:20:19.662 UTC: 247 reports2026-10-03 13:23:57.005 UTC: 248 reports2026-10-03 19:14:07.032 UTC: 245 reports2026-10-03 23:05:36.952 UTC: 242 reports2026-10-04 05:57:54.519 UTC: 244 reports2026-10-04 14:05:19.254 UTC: 243 reports2026-10-04 19:38:32.895 UTC: 236 reports2026-10-04 23:11:48.369 UTC: 235 reports2026-10-05 05:42:52.173 UTC: 236 reports2026-10-05 17:09:31.397 UTC: 221 reports2026-10-06 01:18:29.827 UTC: 222 reports2026-10-06 11:17:28.442 UTC: 206 reports2026-10-06 18:23:24.007 UTC: 200 reports2026-10-06 23:49:20.886 UTC: 202 reports2026-10-07 06:02:10.270 UTC: 204 reports2026-10-07 15:34:02.304 UTC: 199 reports2026-10-07 21:18:58.749 UTC: 199 reports2026-09-30 20:49:27.091 UTC2026-10-07 21:18:58.749 UTC

Each dot is one successful scheduled observation. Counts use a rolling 14-day creation window; reports can leave that window. No measurements are implied between scans.

Exact observation times and counts
  • : 293 reports
  • : 285 reports
  • : 255 reports
  • : 252 reports
  • : 247 reports
  • : 240 reports
  • : 241 reports
  • : 247 reports
  • : 248 reports
  • : 245 reports
  • : 242 reports
  • : 244 reports
  • : 243 reports
  • : 236 reports
  • : 235 reports
  • : 236 reports
  • : 221 reports
  • : 222 reports
  • : 206 reports
  • : 200 reports
  • : 202 reports
  • : 204 reports
  • : 199 reports
  • : 199 reports

Newly observed since the previous scheduled scan

4 reports and 0 topics present now were absent from the previous observation. Newly observed does not mean newly created on GitHub or first ever reported.

Current topics by observed report count

Existing Pulse classifications across the 14-day window. A report can also belong to an emerging topic, so topic counts can overlap.

  1. Usage display & /status

    42 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

  2. Unexpected usage depletion

    38 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

  3. Hitting the limit

    18 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

  4. Requests to change limits

    12 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

  5. Plan changes & allowance

    11 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

  6. “Model at capacity” errors

    9 observed reports

    First seen in scheduled history: .
    Last seen: .
    Present in 41 scheduled observations.

Current reports: source links and observation history (199)

First/last seen refer only to retained scheduled observations. GitHub creation/update times describe the source issue. Absence from a later scan does not prove resolution.

Snapshot updated Oct 7, 2026, 21:18 UTC. Select a topic for its questions and sources.

This week at a glance

Most reported
Unexpected usage depletion
15 reports in 7 days
Fastest rising
No clear riser
No topic grew enough to call it a trend.
New recurring terms
None this week
No unfamiliar name recurred across 3+ reports

What changed: 74 usage-related issues were filed in the last 7 days, against 125 the week before. Overall volume is roughly steady. 12 came in over the last 24 hours.

Topics this week

Ordered by reports in the last 7 days. Open a topic for the actual questions, links to each GitHub issue, and what Codex Usage already covers.

Unexpected usage depletionSteady15 this week · 23 the week before

15 reports in the last 7 days, 23 the week before. Roughly steady. 2 in the last 24 hours.

Models or modes named in these titles: Astra (4), Sol (4), Luna (4), Reasoning effort (2).

What this shows

About as many reports as the week before. This is a steady, recurring question, not a new spike.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

Why Codex usage drains fast

Usage display & /statusSteady15 this week · 27 the week before

15 reports in the last 7 days, 27 the week before. Roughly steady. 1 in the last 24 hours.

What this shows

About as many reports as the week before. This is a steady, recurring question, not a new spike.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

Where to check Codex usage

“Model at capacity” errorsSteady5 this week · 4 the week before

5 reports in the last 7 days, 4 the week before. Roughly steady. None in the last 24 hours.

Models or modes named in these titles: Sol (1), Astra (1), Reasoning effort (1).

What this shows

About as many reports as the week before. This is a steady, recurring question, not a new spike.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

Rate limit and capacity errors

Requests to change limitsCooling4 this week · 8 the week before

4 reports in the last 7 days, down from 8 the week before (-50%). 2 in the last 24 hours.

Models or modes named in these titles: Astra (1).

What this shows

Fewer reports than the week before. That can mean a fix shipped, or just a quieter week. The issue stream can't tell those apart.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

The 5-hour limitThe weekly limit

Hitting the limitCooling4 this week · 14 the week before

4 reports in the last 7 days, down from 14 the week before (-71%). 1 in the last 24 hours.

Models or modes named in these titles: Astra (1), Daybreak (1), Reasoning effort (1).

What this shows

Fewer reports than the week before. That can mean a fix shipped, or just a quieter week. The issue stream can't tell those apart.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

What happens when you hit the limitCodex limits by plan

Plan changes & allowanceCooling3 this week · 8 the week before

3 reports in the last 7 days, down from 8 the week before (-62%). 2 in the last 24 hours.

What this shows

Fewer reports than the week before. That can mean a fix shipped, or just a quieter week. The issue stream can't tell those apart.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

Codex limits by planCodex credits

Banked & instant resetsFew reports3 this week · 3 the week before

3 reports this week, 3 the week before. Too few to call a trend. 1 in the last 24 hours.

What this shows

Only a handful of reports. Treat these as individual accounts, not a pattern.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

Banked resets explainedInstant resets

Token accounting & compactionFew reports3 this week · 3 the week before

3 reports this week, 3 the week before. Too few to call a trend. None in the last 24 hours.

Models or modes named in these titles: Astra (3), Reasoning effort (2), Sol (1).

What this shows

Only a handful of reports. Treat these as individual accounts, not a pattern.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

What counts toward Codex usage

Reset timing & windowsFew reports1 this week · 1 the week before

1 report this week, 1 the week before. Too few to call a trend. None in the last 24 hours.

What this shows

Only a handful of reports. Treat these as individual accounts, not a pattern.

Evidence type: community-reported GitHub issues. Not official, not measured by Codex Usage.

When Codex limits reset

How the Pulse works, and what it can't tell you

  • Source: public issues in the openai/codex repository. Everything OpenAI labels rate-limits, plus issues whose titles clearly mention usage, quota, limits or resets. Titles about CPU usage, context limits or usage policy are left out.
  • Windows: the last 24 hours, the last 7 days, and the 7 days before that, counted back from the update time. Of 229 issues checked in that span, 199 were usage-related.
  • Topics: each issue title is matched against fixed rules, with no AI in the loop. An issue lands in one topic only. 55 usage-related issues didn't fit any topic and aren't shown in one.
  • Emerging terms: a name that shows up in at least three different issue titles, on more than one day, and isn't part of what this site already covers.
  • Trends: a topic is only called rising or cooling when it moves by at least 3 reports and at least half of the previous week. If neither week reaches 5 reports, there's no trend call, and percentages only appear once the previous week has at least 5.
  • Limits: GitHub issues come from people motivated enough to file one. The counts show what's being reported, not how many Codex users are affected, and they don't confirm any change on OpenAI's side. Everything on this page is community-reported evidence.
  • Freshness: successful scheduled scans are persisted about every 4 hours, then published through a site deployment. Failed scans leave the last successful observation in place. The same data is available as JSON.