Will Codex reset today?
We follow posts from @thsottiaux, identify reset signals and send reminders, so you don’t have to keep checking X.
Estimated reset date
Public reset events and intervals do not determine your account’s reset time.
Latest event time
Typical reset interval3.16 days
Source
Event history
Loading local time zone…
Loading…
Loading…
FAQs
Does the public status mean my account has reset?
No. The public status shows announcements and recorded public events with their stated scope and original sources; it does not read your account. Check the affected users and whether an announced action has been confirmed. A reset credit gives eligible users a way to reset usage; receiving one does not automatically restore your allowance. Confirm your actual limits in the official Codex client.
Where can I find my actual remaining allowance?
Use the Codex usage dashboard for your account’s limits and reset times. The CLI /status also reports available usage information. Local tasks and cloud chats share plan usage, and a weekly limit may apply in addition to the five-hour window. See OpenAI’s usage guide. ChatGPT Work and Codex share usage, credits and limits, so also check work done in ChatGPT Work on the same account when investigating unexpected consumption.
Why can similar prompts consume different amounts?
The model, conversation context, reasoning, tool calls, retrieval and caching all affect use. The local message ranges above are estimates, not a fixed allowance per prompt. See OpenAI’s explanation of usage.
Are plan usage, extra credits and an API key interchangeable?
No. Included usage is subject to your plan’s limits; eligible accounts may extend use with credits. An API key uses usage-based API billing and does not refill a subscription window. Check Codex pricing and usage for the current rules.
What happens when my timer reaches zero?
Only your local countdown ends. The timer stores the time you entered in this browser; it cannot observe a Codex reset or change your allowance. Recheck the official usage dashboard before resuming work.
Where can I compare plan limits?
Compare the five-hour message estimates by plan, then confirm the window and model in your own account. For how reasoning, tools and subagents affect consumption, read how Codex usage is counted.
Models & usage
Latest models and allowance use
Checked 16 Oct 2026Compare like for like: token rates describe a fixed amount of input and output, while a real task can use very different amounts of both.
| Model | Best for | Input / output per 1M tokens | Plus local messages / 5 hours (estimate) |
|---|---|---|---|
| GPT-6 Astra | Hard end-to-end work | 250 / 1,250 credits | 5–45 |
| GPT-6 Sol | Complex coding and agent tasks | 50 / 250 credits | 15–150 |
| GPT-6 Luna | Focused, high-volume tasks | 2.5 / 12.5 credits | 350–3,000 |
Input / output per 1M tokens: Codex Standard credit rates are 100:20:1 for Astra, Sol and Luna at the same token volume. Included plan usage is not a fixed number of messages; Fast uses included plan allowance at 2.5× Standard, while purchased credits and Enterprise pay-as-you-go use 2× Standard where available. Check your own usage dashboard for remaining allowance and reset times. Model availability depends on your account, client and rollout.
Official pricing and usage: OpenAI Codex pricing · GPT-6 rollout
Plan limits
Codex limits by plan
ReviewedEstimated local messages per five-hour window for these models. Ranges vary with the work; they are not fixed message quotas or daily allowances. Model access depends on your plan, rollout and workspace settings.
| Model | Plus | Pro 5x | Pro 20x | Standard Business |
|---|---|---|---|---|
| GPT-6 Astra | 5–45 | 25–225 | 100–900 | 5–45 |
| GPT-6 Sol | 15–150 | 70–700 | 300–3,000 | 15–150 |
| GPT-6 Luna | 350–3,000 | 1,750–14,000 | 7,000–56,000 | 350–3,000 |
Local and cloud work share plan usage; weekly limits may also apply. ChatGPT Work shares Codex usage. Standard Business uses the Plus estimates; Business ($100) uses Pro 5x estimates. Check the affected window in your account instead of multiplying this table into a weekly allowance.
Other plans and billing
- Free / Go: the current pricing page lists GPT-6 Luna at Standard speed in the desktop app, subject to rollout. It does not give a comparable message range here; check your account.
- Enterprise / Edu: flexible pricing uses credits without fixed rate limits in this reference. Without flexible pricing, most features follow Plus per-seat limits. Workspace controls and your agreement still apply.
- API key: separate usage-based API billing and model/tier rate limits. Subscription message estimates do not apply. The credit rates in the model comparison above describe ChatGPT credit use, not API dollars.
Official source · Check your limits and reset times
Usage rule changes
Dated official changes relevant to usage. A rollout announcement does not establish when every account received access; public reset events remain in the history above.
GPT-6 Sol and Luna rollout
Announced
The announcement compares the new models with their GPT-5.6 predecessors and reports lower token prices. It covers Plus, Pro, Business, Enterprise and Edu; Free and Go get Luna in the desktop app. Access rolls out and Enterprise admins must enable the models. It gives no universal completion date or full historical price table.
Official sourceAdditional credits for Plus and Pro
Announced
The announcement adds on-demand credits for Codex use beyond the included plan allowance. It describes availability at announcement, without a separate activation time or a record of earlier continuation options. Buying credits is not an account reset.
Official source
Guides
Codex
Codex 429 Too Many Requests: Retry Limit and Error Fixes
Fix Codex exceeded retry limit, last status: 429 Too Many Requests. Distinguish request throttling, API quota, sign-in errors and interrupted streams.
Last checked · Read guideCodexCodex Limit Reached: Check Your Account and Resume Work
Codex limit reached or still blocked after a reset? Check your account, five-hour and weekly windows, earned resets, and the balance needed to resume work.
Last checked · Read guideCodexCodex Reset Time: 5-Hour vs Weekly Limits
Why did your Codex five-hour allowance reset but not your weekly limit? Learn how to check /status, avoid limit traps, and track your actual Codex reset time.
Last checked · Read guideCodexHow Codex Usage Works: Models, Reasoning, Tools and Subagents
Understand why a short Codex prompt can use more allowance: compare models, reasoning, tool output, caching and subagents without confusing credits with plan limits.
Last checked · Read guide