ChatGPT limits aren't one counter: everyday text chat on Free is described as unlimited subject to abuse-prevention safeguards, while uploads, images, voice and data analysis have separate limits. As of Nov 2026 OpenAI says that starting in December, Plus, Pro and Business usage allowance will be shared across ChatGPT and Codex and measured in tokens.
So when ChatGPT stops you, the first question isn't "what's my limit?" but "which limit did I hit?" Here's how to tell them apart, using only OpenAI's own help pages.
Five different things can run out
- A tool limit. File and image uploads, image creation, voice and data analysis each have their own limits on Free.
- A model limit. A model or intelligence option is still shown but greyed out, with a message or a reset time.
- Your plan's allowance. From December, one combined allowance for ChatGPT and Codex on Plus, Pro and Business.
- The conversation itself. "Context window full" or "Chat memory full", which isn't a usage limit at all.
- An account restriction. Access limited for security, policy or billing reasons.
Each one has a different fix, which is why it pays to read the message before you do anything else.
What Free gets
OpenAI's Free tier FAQ says Free users have unlimited everyday text chats, subject to abuse-prevention safeguards. File uploads, image generation, voice, data analysis and other tools have separate usage limits, and ChatGPT tells you when you reach one. If you hit a tool limit, you'll need to wait for a later time to use it again.
GPTs work a little differently. GPT usage follows your current Free tier model limit, and when you reach it, GPT access pauses until the limit resets. ChatGPT shows the reset timing in the product.
Two more things worth knowing. If you upgrade to Plus, Pro or Business after hitting a Free limit, OpenAI says your usage rates reset. And the default model and the limits themselves can change over time, so treat anything you read, including this page, as a snapshot. For the wider question of what you get without paying, see whether ChatGPT is free.
The December change for paid plans
OpenAI's usage limits page says: "Starting in December, Plus, Pro, and Business plans' usage allowance will be shared across ChatGPT and Codex as we bring the agentic capabilities of Work into the main ChatGPT experience."
Usage will be measured in tokens, the units of information a model reads and writes. How much of your allowance you use depends on the model you choose and the task you give it. More advanced models and complex tasks generally use more. And one line that's easy to miss: "Instant chat will not draw from your usage."
OpenAI says it expects most people won't need to wait for a reset more often than they do now. That's a prediction, not a promise, so I'd watch the first few weeks. If you're deciding between paid plans, my Plus vs Pro comparison covers the differences.
Where to see your usage and reset time
OpenAI says you'll see your combined remaining usage and reset times in Settings, then Usage. If you reach the limit, its page gives three options: wait for the reset, upgrade your plan, or buy credits to keep using eligible features.
If a model or option is shown but disabled, read the message next to it. If it says you've reached a limit or shows a reset time, wait until then, or pick another available option.
| What ran out | What OpenAI's page says | What to do |
|---|---|---|
| A Free tool limit (uploads, images, data analysis) | These tools have limits separate from chat; ChatGPT notifies you | Wait for a later time, or upgrade, which resets Free limits |
| GPT access on Free | Follows your Free tier model limit and pauses until it resets | Check the reset time shown in the product |
| The chat is full | It's about how much the conversation can hold; there's no fixed message count | Start a new chat and bring over only what you need |
| A model option is greyed out | Could be capacity, a usage limit or workspace permissions | Read the message, try later, or choose another model |
| Your plan allowance, from December | Shared across ChatGPT and Codex, measured in tokens | Wait for the reset, upgrade, or buy credits |
| Access limited after login trouble | Suspicious activity can prompt a temporary downgrade | Reset your password and turn on two-factor authentication |
When ChatGPT says the chat is full
This one confuses people most, because it looks like a limit and isn't. OpenAI says "Context window full" and "Chat memory full" refer to how much information ChatGPT can use in the current conversation. Your messages and its replies both take up space. A few long exchanges can fill it faster than many short ones, so there's no fixed number of messages per chat.
The limit applies to that one conversation. You don't need to turn off Memory, delete saved memories or clear other chats. On Free you can start a new chat straight away. If the notice offers a new chat with a summary, that summary can carry some context over, but it may drop an earlier instruction or your exact wording. If you're working on a draft, paste the passage you want to revise and tell ChatGPT where to pick up.
You can still read the original chat even when you can't send another message in it. And if a brand new chat says it's full after one short message, OpenAI suggests checking OpenAI Status for an incident, then trying the same message in a private browser window.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Why a model option is greyed out
OpenAI's troubleshooting page lists six reasons model access may be limited: the model is at capacity; workspace permissions in a managed workspace; a usage limit; a possible Terms of Service flag; security concerns such as several failed logins or access from unknown locations; and recent changes to your subscription, plan or payment details.
If you're in a company workspace, check you're in the right workspace first, then ask a workspace owner to review your access. OpenAI points out that workspace permissions are separate from usage limits, so a model a colleague can use might not be available to you.
Credits
Credits let you keep using eligible features after your plan's included limits run out. OpenAI says your plan usage is used first, then credits, and "Nothing about your plan changes." Supported features include Codex, ChatGPT Work, Word, Excel and PowerPoint, depending on your plan.
You can view your balance and buy credits in Settings, then Usage, where available. Automatic reload is optional and off unless you turn it on; if your balance is already below your chosen minimum when you switch it on, a purchase may happen straight away. Credits last 12 months from purchase and aren't API credits. For Free and Go, OpenAI says credit purchases are available to a limited group so far. Check OpenAI's own page for current rates, and my page on ChatGPT credits for how they work.
Steps when you hit a wall
- Open Settings, then Usage, and see what's left and when it resets.
- If a greyed-out option shows a reset time, wait until then or pick another available model.
- If the message says the chat is full, start a new chat and paste in only the part you're working on.
- If a model is at capacity, try again in a few minutes or switch model, and check the OpenAI status page.
- In a company workspace, confirm you're in the right one and ask the owner to check your role.
- If access was restricted after login trouble, reset your password and turn on two-factor authentication, then contact OpenAI Support if it doesn't come back.
What I'd do
I'd check Settings, then Usage, before blaming the plan. Half the time the message is about the conversation, not the account. For long jobs, I'd start a fresh chat early and paste in only what's needed, rather than dragging a huge thread along.
I'd skip credits until I've seen how fast a normal working week uses the new combined allowance. If you also use Google's assistant, it's worth seeing how Gemini handles its limits, since the two work quite differently.
Frequently asked questions
How many messages can I send on ChatGPT Free?
OpenAI doesn't give a number for everyday text chat. It says Free users have unlimited everyday text chats, subject to abuse-prevention safeguards, while uploads, image creation, voice and data analysis have separate limits.
When does my ChatGPT limit reset?
ChatGPT shows the reset timing in the product when you hit a limit. OpenAI says you'll see combined remaining usage and reset times in Settings, then Usage.
Does upgrading reset my limit?
Yes. OpenAI says if you reach a rate limit on Free, your usage rates reset when you upgrade to a Plus, Pro or Business plan.
Do I need to delete memories when ChatGPT says the chat is full?
No. OpenAI says the full-chat limit applies to that conversation only. You don't need to turn off Memory, delete saved memories or clear other chats. Start a new chat instead.
Do credits change my plan?
No. OpenAI says nothing about your plan changes. Your included usage is used first, and credits only cover eligible features after that.
