# Codex Rate Limits Discussion Thread

**URL:** <https://community.openai.com/t/codex-rate-limits-discussion-thread/1378553>\
**Category:** Codex\
**Tags:** rate-limit\
**Created:** [April 4, 2026, 10:54pm UTC](https://community.openai.com/t/codex-rate-limits-discussion-thread/1378553 "2026-04-04T22:54:52Z")\
**Posts on this page:** 1\
**Showing post:** 107

<div class="post-metadata">

**Author:** ![Mat\_Tr](https://avatars.discourse-cdn.com/v4/letter/m/c37758/32.png) [@Mat\_Tr](https://community.openai.com/u/Mat_Tr)\
**Post date:** [April 6, 2026, 7:57pm UTC](https://community.openai.com/t/codex-rate-limits-discussion-thread/1378553/107 "2026-04-06T19:57:27Z")

</div>

I think the issue here is not just that the promo ended and limits were reduced.

The problem is how usage is being valued.

Look at a simple example:

— 30k tokens = ~14% of the 5h limit  
That already looks too expensive and unrealistic.

Now compare it to API pricing:

— 1M tokens ≈ $2–3  
— A $20 subscription ≈ ~10M tokens via API

But with current Codex limits:

— 1,000 tokens ≈ 0.05% of the week limit  
— So 10M tokens ≈ ~500% of week limit

That means to use the same 10M tokens you would need:  
— 5 accounts ($100 total)  
— and also wait for limits to reset

While via API it costs ~$20 with no waiting.

This doesn’t add up.

A subscription cannot be 5–6x worse than API pricing — by design it should be more cost-efficient, not less.

So this looks less like “reduced limits” and more like an issue with how token usage is being priced/calculated internally.

---

_[View the full topic](https://community.openai.com/t/codex-rate-limits-discussion-thread/1378553)._