ChatGPT "You're Making Requests Too Quickly": What It Means and Fixes

If ChatGPT shows “Too many requests – You’re making requests too quickly”, it has temporarily blocked your browser or app from loading your conversations, and the fix in almost every case is to stop clicking, wait a few minutes, and try again. It is not your message allowance running out and it is not a ban, but it can recur if whatever triggered it (heavy sidebar scrolling, bulk archiving, many tabs or devices) keeps happening.

Below is what the error actually is, what seems to set it off, and the steps that have worked for other users as of November 2026. One caveat up front: OpenAI has not published an official explanation or fix for this specific message, so much of what follows comes from detailed user reports on OpenAI’s own GitHub issue tracker and community forum.

What the full message says

The modal reads, word for word, as reported in issue #38503 and issue #46475:

Too many requests You’re making requests too quickly. We’ve temporarily limited access to your conversations to protect your data. Please wait a few minutes before trying again.

Variants people have reported include:

• Desktop app: “Request failed with status 429”
• Web: a conversation that won’t open with “Could not load this ChatGPT conversation”
• Android: a short orange “Too many requests” toast

All of these point to the same thing: an HTTP 429 response, the standard “slow down” code a server sends when one client is making too many requests.

Why it happens

It’s about loading chat history, not sending prompts

The key phrase is “limited access to your conversations.” Users who dug into their browser’s network tools found that the failing request is the one that fetches your chat list. In issue #38763, a single request to the conversations list endpoint returned 429 right after opening a new chat, and in issue #37518 the desktop app’s /conversations request was the one being rejected. Reports on the OpenAI Developer Community describe ordinary chat requests going through while only the sidebar history load fails.

That’s why paying users see it too. Reports in these threads come from Plus users and several Pro users.

Reported triggers

No single cause has been confirmed, but these come up repeatedly:

One technical detail matters for what you do next: the #37518 reporter found that the client immediately retrying failed requests amplifies the rate limit. In plain terms, hammering refresh likely keeps you locked out longer.

The desktop-app version of the problem was reported to have started after an automatic update on August 7, 2026, and has since been reported on macOS, Windows, Linux and Android, in Chrome, Edge and the desktop app.

How to fix it

Work through these in order. The first two resolve most cases.

1. Stop and wait. Close the error, don’t refresh, and leave ChatGPT alone for at least a few minutes. Users report blocks of roughly 10–20 minutes; a few say it lasted until the next morning.
2. Close duplicate sessions. Close extra ChatGPT tabs, and if you have both the desktop app and the website open, quit one. A user in #37518 suspected the two were loading history at the same time.
3. Avoid the sidebar for a while. When access returns, open chats via search or a direct link rather than scrolling far down the history list.
4. Slow down bulk actions. If you’re archiving, unarchiving or deleting lots of chats, do a few at a time with pauses in between. There is no official bulk “unarchive all” option at the moment, per #46475.
5. Test without extensions. Open ChatGPT in a private window with extensions disabled, or turn off any ChatGPT-related extensions. One user’s problem disappeared after this, though they couldn’t confirm the extension was the cause.
6. Sign out of other devices. In ChatGPT, go to Settings → Security and use the option to log out of all devices, then sign back in on the one you’re using. One user with 28 active devices said this stopped the errors for a full day; another said it didn’t help, so treat it as worth a try rather than a guaranteed fix.
7. Check OpenAI’s status page. If none of the above works, look at status.openai.com for an active incident. If one is listed, waiting it out is the only option.

Fixes that probably won’t help

• Rapid refreshing or reopening the app sends more of the same requests that caused the block.
• Upgrading your plan isn’t a fix here. Pro and Plus subscribers report the same error.
• VPNs, “IP reset” tricks or third-party tools aren’t supported by any of the evidence and may add new problems. Stick to the official app and website.

How it differs from other ChatGPT limits

It’s easy to confuse this message with other limits:

• Plan usage limits cap how many messages you can send to certain models in a time window. Those messages are about your plan’s allowance, not about conversations being “temporarily limited to protect your data.”
• API 429 errors affect developers calling OpenAI’s API. They’re tied to an organization’s requests or tokens per minute and are handled with retry-and-backoff logic in code, which is separate from the consumer ChatGPT app.

If your message mentions conversations and asks you to wait a few minutes, you’re dealing with the history-loading rate limit described in this guide.

Is OpenAI fixing it?

As of November 2026, the main GitHub reports (#38503, #38763, #37518, #46475) are still open and carry a “rate-limits” label, but none show a reply from OpenAI staff, only automated labelling and duplicate detection. New reports were still being added through late October 2026.

That means the behaviour could change at any time, either because OpenAI loosens the limit or because a client update stops the app from firing so many history requests. If you keep hitting it during normal use, adding your details (platform, app version, what you were doing) to the existing reports is the most useful official channel, rather than opening a new duplicate.

The short version

“You’re making requests too quickly” means ChatGPT temporarily stopped your app from loading conversations because it saw too many history requests in a short time. Wait a few minutes without refreshing, close extra tabs and duplicate apps, go easy on sidebar scrolling and bulk archiving, and, if it keeps coming back, sign out of old devices and check for extension interference.