openai / openai/codex

Feature request: Let users choose when the 5-hour Codex usage window starts

Open
#41,089 3 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement rate-limits
Dominant language
Rust
Stars
125k
Forks
19.5k
PR merge metrics
PR metrics pending

Description

Summary

Please let users control when their Codex 5-hour usage window starts instead of having the window begin automatically from the first small use.

This is not primarily a request for more quota. It is a request for predictable quota timing so users can plan serious coding sessions around the existing limit.

Problem

The current rolling / user-triggered 5-hour window is difficult to plan around.

A tiny Codex use at an inconvenient time can determine the reset timing for the next several hours. Later, when the user starts a substantial autonomous task, a single task can consume a large fraction of the 5-hour allowance and the session may be interrupted in the middle.

For long coding sessions, the main problem is not only how much quota exists, but when the window starts and when it will become available again.

Proposal: user-controlled start button

Add a control such as:

Start 5-hour window

Suggested behavior:

  • The 5-hour Codex allowance does not become active until the user starts the window.
  • Pressing the button starts the user's 5-hour window.
  • The button then enters a strict 5-hour cooldown and cannot be used again until the cooldown expires.
  • Show the exact next-available time clearly.
  • Optionally provide Auto Start for users who prefer today's automatic behavior.
  • Ideally also allow a scheduled start time, e.g. Start every day / cycle at 10:00 PM.

Example:

10:00 PM
[Start 5-hour window]

Window active
Next window available: 3:00 AM

The user cannot restart the window early, so this does not require removing the existing rate-limit protection.

Why this is better

This would let users deliberately align Codex capacity with their actual work schedule.

For example, if I finish normal work at 6 PM but plan to do serious Codex development at 10 PM, I should be able to save the start of my 5-hour window until 10 PM rather than accidentally starting it with a small request earlier in the evening.

For users who want no extra management, Auto Start could preserve the current experience.

Benefits

  • predictable reset timing;
  • easier planning for long autonomous tasks;
  • less chance that a trivial request starts an inconvenient quota window;
  • users can see whether they have enough capacity before beginning a large task;
  • does not necessarily require increasing total weekly compute;
  • preserves a strict 5-hour cooldown between window starts.

Related issues

  • #40725 proposes replacing the 5-hour window with a fixed local-time daily quota.
  • #40905 describes long-running tasks being interrupted by the 5-hour limit.
  • #28931 requests automatic resume after the 5-hour limit resets.

This proposal is a narrower alternative: keep the 5-hour model, but give users control over when the window begins, with an optional automatic/scheduled mode.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or implementation entry points are identified in the issue. Start by locating the Codex usage-window and quota-handling entry points, then clarify whether manual, automatic, and scheduled starts are all in scope; done should include a strict five-hour cooldown and a clearly reported next-available time.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
cli
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.