# Codex Reset — full text for language models

> A free, independent tracker for OpenAI Codex usage-limit resets: verified reset history, a live probability that another reset lands in the next 24–48 hours, and browser-only countdowns for your own 5-hour and weekly windows.

Canonical site: https://codex-reset.com/ (formerly codexreset.net)
Site version: 1.7.0.0
Content snapshot: 2026-08-03
Contact: hello@codex-reset.com

## What this is

Independent community project, not affiliated with OpenAI. "Codex" and "ChatGPT" are trademarks of OpenAI and are used here only to describe what is being tracked.

No account, no sign-up, no paywall. Personal timers and preferences are stored in the visitor's browser and are never uploaded.

The interface is published in English, Chinese (zh-Hans), Japanese and Spanish; this file and the release notes are English only.

## What this site does not claim

- This site cannot see anyone's OpenAI account, quota, or usage. The countdowns are driven entirely by times the visitor enters, and the paste analyzer on /codex-usage parses text the visitor supplies without sending it anywhere.
- A reset "forecast" is a probability derived from the cadence of past verified announcements. It is not an OpenAI schedule, a leak, or a commitment, and this site does not predict outages.
- A reset is called verified only when a public announcement says so. Timing that merely sits near an incident is labeled temporal, never causal.
- The Codex limit numbers on this site are quoted from OpenAI's published pages with the date they were last checked. Where OpenAI has published nothing — the size of the weekly window, for example — this site says so instead of estimating.

## Pages

### Reset forecast — https://codex-reset.com/

Answers "will Codex reset today?" with the last verified global reset, a 24h/48h probability, and personal 5-hour, weekly and banked-reset countdowns that stay in local storage.

### Reset timeline — https://codex-reset.com/timeline

Every verified reset, boost, unlock and banked-reset announcement, with timezone-aware charts, filters and a sortable table.

### Signal radar — https://codex-reset.com/radar

Live model configuration, OpenAI service health, reset events and limit signals in one evidence-labeled feed.

### Usage workbench — https://codex-reset.com/codex-usage

What every block of OpenAI's Codex usage dashboard means, plus a browser-only analyzer: paste your `codex /status` output and get your weekly window, burn pace and projected reset back.

### Model & effort selector — https://codex-reset.com/codex-juice

Which Codex model (Sol, Terra, Luna) and which Low–XHigh reasoning effort fits a task, with the public evidence behind each recommendation.

### Status & compensation — https://codex-reset.com/status

Live Codex service state, official incident history, public recovery times, and which resets OpenAI explicitly linked to an outage.

### Tibo radar — https://codex-reset.com/tibo

Verified @thsottiaux signals, the reset announcement rail, and the community reset vigil.

### Atom feed — https://codex-reset.com/feed.xml

Verified announcements as a standard Atom feed, for readers and crawlers.

Localized routes: the pages above are also published under /zh, /ja and /es, except /feed.xml, /codexreset-release, /terms and /privacy, which exist once in English.

## Alert channels

- Telegram (@codexresetalerts, live): Public channel; one message per verified reset signal.
- Discord (#alerts, live): The same verified signals relayed into the community server.
- Browser push (alerts hub, live): Payload-free Web Push for verified resets — no account, nothing personal in the message.

Live channels carry the same verified reset signals. Nothing is sent for an unconfirmed rumor.

## Public JSON endpoints

Public GET responses require no key and allow cross-origin reads. /api/pray also accepts the bounded interactions exposed on /tibo.

### https://codex-reset.com/api/forecast

Current reset probability for the next 24h and 48h, the confidence band behind it, the last verified reset timestamp, and the cadence samples the model used.

### https://codex-reset.com/api/timeline

Every verified reset, boost, unlock and credit announcement with its date, type, summary and source URL.

### https://codex-reset.com/api/status-history

Current Codex service state per surface plus the 90-day incident record and any reset explicitly linked to an outage.

### https://codex-reset.com/api/feed

The announcement feed as ingested, with fetch time, content age, staleness flag and the current headline signal.

### https://codex-reset.com/api/juice

Observed Codex model and reasoning-effort data with per-sweep freshness, coverage and verification state.

### https://codex-reset.com/api/usage-reference

An anonymous day-by-day reference curve of percent-remaining across a reset cycle.

### https://codex-reset.com/api/pray

GET returns aggregate counts for the community reset vigil on /tibo; POST records the page's bounded community interactions.

Announcements are also published as Atom at https://codex-reset.com/feed.xml.

## Verified reset record

These counts describe the curated baseline compiled into this content snapshot. The baseline covers dated announcements through 2026-07-28. Announcements after that coverage date are served live and are not in this file — read https://codex-reset.com/api/timeline for the current list.

- Announcements in the baseline: 39, dated 2025-09-17 through 2026-07-28.
- Reset announcements: 31 — 25 confirmed as already in effect, 6 announced with an official forward window such as "landing within the hour".
- Limit boosts: 5. Model or feature unlocks: 1. Credit grants: 2.
- Longest gap between verified resets: 73 days. Recent pace: about one every 2 days.

## Codex limit reference

Quoted from OpenAI's published pages, last checked 2026-07-25. Source: OpenAI Help Center: Using Codex with your ChatGPT plan — https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan.

### the 5-hour window

The short clock. Local messages (CLI, IDE extension, desktop app) and cloud chats draw from one shared 5-hour budget — they are not separate allowances. When your Codex CLI reports this window, /status shows its percentage left and reset time; use the rows your own client actually prints.

### the weekly window

The ceiling behind it, running seven days. OpenAI has never published how large it is for any plan — the official pricing page says only that "additional weekly limits may apply". Clearing a 5-hour window does not restore weekly budget; the two refill independently.

### what a reset actually is

The reset time is not computed on your machine. OpenAI's servers send an absolute resets_at timestamp with each usage report, and the CLI just renders it. A reset restores your percentage to 100% — it doesn't add credits or change your plan. Community reports suggest each window's clock starts on your first request after a reset, though OpenAI hasn't confirmed that.

### goodwill & banked resets

Separately from the scheduled windows, OpenAI keeps resetting everyone's usage as a gesture — 37 recorded times since September 2025, announced almost only on @thsottiaux's X feed. Banked resets (since June 2026) sit in your account until you spend them, and expire 30 days after they're granted.

### 5-hour limit status

OpenAI temporarily lifted the 5-hour restriction for Plus, Pro and Business on Jul 12, 2026 during the GPT-5.6 surge, and last confirmed it was still lifted on Jul 14. No end date was announced, and the pricing page still documents 5-hour windows — so treat your own /status output as the live truth.

### plan-level summary

| plan | codex access | when you hit the wall |
| --- | --- | --- |
| Free / Go | included for quick, lightweight tasks — no published numbers | wait for the window to roll over |
| ChatGPT Plus $20 | full GPT-5.6 family across CLI, IDE, web, cloud | buy credits (Settings → Usage → Credits) or wait |
| ChatGPT Pro $100 / $200 | 5× or 20× the Plus windows, plus research previews | credits with auto top-up, or wait |
| Business | same per-seat windows as Plus + admin controls | workspace credits |
| Enterprise / Edu | flexible pricing = no fixed rate limits; usage scales with credits | credits pool |
| OpenAI API key | pay per token — no 5h or weekly windows at all | only your own spend limit (loses cloud features) |

Codex, ChatGPT Work, ChatGPT for Excel and Workspace Agents share one agentic usage pool; ordinary ChatGPT chat is metered separately and never drains it. Credits work out to $0.04 each, average GPT-5.6 message ≈ 5–40 credits, and purchased credits expire after 12 months.

## Release history

Versions read MAJOR.MINOR.PATCH.BUILD. Dates are production deploys.

### 1.7.0.0 — Release notes (2026-08-03)

- new: Public release notes at /codexreset-release: the version history plus a plain-language map of what every page, alert channel and public API does.

### 1.6.0.0 — Alerts hub (2026-08-03)

- new: The nav bell now opens one alerts hub holding every channel — browser push, the Telegram alert channel and the Discord #alerts stream — instead of linking a single one.
- improved: The browser notification permission has a single owner shared by the tracker alarm, the /codex-usage alarm and push, so one prompt covers all three.

### 1.5.0.0 — Codex usage workbench (2026-08-02)

- new: /codex-usage rebuilt as a two-pane workbench: paste your `codex /status` output and read your current cycle back — weekly window, burn pace and projected reset. The paste never leaves the browser.
- new: A community reset forecast poll on the same page; results unlock once five forecasts are in.
- improved: The paste parser reads OpenAI's boxed `/status` layout, including wrapped lines and dated resets such as "12:35 on 8 Aug".

### 1.4.1.0 — Browser push notifications (2026-08-01) [live]

- new: Web Push for verified resets, built payload-free with VAPID: the push message carries no content, and the browser fetches the details from this site when it wakes. No account, no email, nothing personal stored with the subscription.
- improved: Notifications fire only for a reset this site has verified, never for an unconfirmed rumor.

### 1.4.0.0 — Discord alerts and Atom feed (2026-08-01)

- new: Verified reset announcements relay into the community Discord #alerts channel, with a permanent invite in the alerts dock and the footer.
- new: An Atom feed at /feed.xml, discoverable from every page.

### 1.3.0.0 — Codex Juice (2026-07-31)

- new: /codex-juice: a task-first selector for the Codex model (Sol, Terra, Luna) and the Low–XHigh reasoning effort, published with the benchmark evidence it rests on.
- new: The same evidence as public JSON at /api/juice.

### 1.2.0.0 — Tibo radar and Telegram alerts (2026-07-28)

- new: /tibo tracks verified @thsottiaux signals and reset announcements on a card rail, alongside the community reset vigil.
- new: The @codexresetalerts Telegram channel posts one message per verified reset signal.

### 1.1.0.0 — Four languages (2026-07-26)

- new: English, 中文, 日本語 and Español, each with its own routes, hreflang matrix, share cards and localized public APIs.

### 1.0.0.0 — Launch (2026-07-25)

- new: codex-reset.com goes live, moving the project off codexreset.net: the reset forecast, personal 5-hour and weekly countdowns, /timeline, /radar and /status.
- new: Public read-only JSON from the start — /api/forecast, /api/timeline, /api/feed and /api/status-history — so the record stays checkable outside this site.

## Where the data comes from

- Reset announcements come from public posts by OpenAI staff, primarily @thsottiaux, ingested on a short delay and stored with the source URL and announcement timestamp.
- Service state and incident history come from OpenAI's public status page, backfilled 90 days and retained 180 days forward.
- Model and reasoning-effort observations on /codex-juice come from a scheduled collector; each row carries its own observation time and freshness state.
- Live reset, forecast and service-status claims resolve to the public JSON endpoints below, where their source and observation times can be checked.

## How to cite this site

- Cite the page URL, not this file, and name the site as "Codex Reset (codex-reset.com)".
- Live values — the current probability, the latest verified reset, current service state — change continuously. Read them from the JSON endpoints rather than repeating a number captured from a page snapshot.
- This file is regenerated on every deploy. The curated reset section prints its own coverage date; anything newer lives in /api/timeline and /api/forecast.