Claude and ChatGPT usage limits: each plan's caps, resets, and what to do at the limit
The Morse teamPublished Facts checked 11 min read
Claude's paid plans have two usage limits, a session limit that refills five hours after your first message and a weekly limit that resets at a fixed time Anthropic assigns to your account, and Anthropic publishes no message count for either. ChatGPT is set up the other way round. OpenAI calls everyday text chats unlimited on every plan, "subject to abuse guardrails," and puts its caps on Thinking models, GPT-6 Pro, deep research, uploads, images, and Codex instead.
We read Anthropic's pricing page and its help articles on the Free, Pro, Max, and Ultra plans, usage limits, limit resets, usage credits, and Fable, then OpenAI's pricing page and its help articles on Free, Go, Plus, Pro, model limits, Codex, deep research, and credits, all on November 20, 2026. Every cap below is quoted as the company states it, and where a page gives no number, we say so. Both companies show the reset time on your own account in Settings > Usage, and OpenAI goes a step further for Codex by publishing estimated message ranges, the only numbers either company puts on a limit.
If you'd rather not plan your day around a meter, Morse is a private AI chat app with GPT, Claude, Gemini, DeepSeek, and uncensored models, encrypted so even Morse can't read your chats. Morse Pro is unlimited messages with no daily cap, for $19 a month or $180 a year. Your history is stored on your device, encrypted, and our servers keep no log of your messages. Every message passes through the Morse Enclave, so the model provider never learns your name, email, account, or IP address.
Claude and ChatGPT usage limits by plan
As of November 20, 2026, in each company's own words, from the pages listed under Sources. "No number" means the company's pages give none.
| Plan | Price | How the company describes the limit | Number given |
|---|---|---|---|
| Claude Free | $0 | "Limited usage every 5 hours and every week." The Free help page says messages "vary based on demand" | No number |
| Claude Pro | $20 a month, or $200 a year | "More usage than Free, every 5 hours and every week." Fable runs on usage credits | No number |
| Claude Max | $100 a month | "5x Pro usage every 5 hours and every week," with Fable "up to half your weekly limit" | A multiple of Pro, no number |
| Claude Ultra | $200 a month | "10x Pro usage every week," and "No 5-hour limit" once the account is verified | A multiple of Pro, no number |
| ChatGPT Free | $0 | "Unlimited text chats," with "Limited" uploads, image creation, voice, deep research, and Codex | No number |
| ChatGPT Go | $8 a month | Unlimited everyday text chats and "More messages with tools." "This plan may include ads." | No number |
| ChatGPT Plus | $20 a month | "Expanded messages and uploads," "Limits apply." In Codex, an estimated 5 to 45 GPT-6 Astra local messages per five hours | Codex estimates only |
| ChatGPT Pro | $100, $200, or $500 a month | "Unlimited subject to abuse guardrails," but GPT-6 Pro has a weekly limit, and Work and Codex an allowance | No number |
The pages don't always agree with each other. Anthropic's pricing page lists a weekly limit on Free, while its Free help page mentions only "a session-based usage limit that will reset every five hours." OpenAI's pricing page offers Free "Unlimited text chats with GPT-6," while its Free tier FAQ says "Free users have access to GPT-5.6 Luna." Where they differ, your own Settings > Usage page is the one to trust.
How does the Claude 5-hour limit work?
The five-hour window opens with your first message and refills completely five hours later. Anthropic's usage article says "The period starts when you send your first message, and your limit resets in full five hours later." If you use it up before then, you wait for the refill or pay to continue (see below). Under Current session in Settings > Usage, you can see how much of the window you've used and how long is left.
Everything you do with Claude draws on that one meter. Anthropic says chats on claude.ai, Claude Code, and Claude Desktop all count toward the same usage limit, and that longer agentic tasks that search the web, run code, or create files "generally use more than a quick question." Message length, attachments, the length of the conversation, the model, and the effort level you pick all change how fast it drains.
Ultra is the exception. Verified Ultra accounts have no five-hour limit, and Anthropic verifies them automatically "using factors like your account and payment history." Until then a temporary limit applies, and verifying a phone number at checkout doesn't remove it. If Settings > Usage shows no Current session bar, your account is verified.
When does the Claude weekly limit reset?
At the same day and time every week, set by Anthropic for your account. Its Pro page says that time stays put "regardless of when you start using Claude or when your subscription begins," and Settings > Usage shows when the next one comes. The two windows run side by side, so a fresh five-hour session doesn't help once the weekly limit is spent, and weekly allowance left over doesn't help inside a spent session.
Max and Ultra carry a third meter, for Fable, Anthropic's largest model family. On those plans you can spend "up to 50% of your weekly usage limits on Fable models at no extra cost," and Anthropic warns that Fable uses the weekly limit up faster than other Claude models. On Pro, Fable sits outside the plan's limits and is paid with usage credits from the first message. A launch offer that put Fable inside Pro's weekly limit ran from November 9 to November 16, 2026, and has ended.
What happens when you hit a Claude usage limit?
You can wait for the reset, upgrade, or turn on usage credits. Anthropic says that after an upgrade "your new plan's limits apply right away," and that with usage credits enabled, "your work continues on usage credits instead of stopping."
Usage credits are pay-as-you-go at standard API rates, billed separately from your subscription. You turn them on in Settings > Usage, buy a balance, and set a monthly spend limit or auto-reload. When your plan's allowance runs out, the warning switches to "Now using credits," and Claude keeps going until you reach your spend limit or empty the balance. Credits bought on or after December 7, 2026 expire 12 months after purchase, earlier ones don't expire, and Anthropic calls purchases "non-refundable" outside what local law requires.
Anthropic also gives eligible plans an occasional limit reset. Clicking Reset for free, in Settings > Usage or on the message you see at a limit, refills the five-hour or weekly limit at once, and it can't be undone. And Anthropic keeps room to cut usage further: its plan pages say it may limit usage "at our discretion" and add "We don't guarantee uninterrupted service." The other places people go when the meters run out are in Claude alternatives.
Is ChatGPT Plus unlimited?
For everyday text chats, yes, and so is every other ChatGPT plan, Free included, "subject to abuse guardrails." The limits sit on everything else. OpenAI's Plus page says the plan "may include usage limits such as message caps, especially during high demand," and "These limits may vary based on system conditions." None of OpenAI's pages gives a number for any cap in Chat.
When you reach a Thinking limit, OpenAI says "ChatGPT may continue with another available model," and it shows when the allowance resets "when that information is available." Pro raises the caps without removing them: "A Pro subscription does not include unlimited use of the GPT-6 Pro model." On Pro $200, reaching the weekly GPT-6 Pro limit moves you to GPT-5.6 Thinking at Medium. On Pro $100, GPT-6 Pro and GPT-5.6 Sol Pro draw from one weekly allowance, so switching between them doesn't buy more.
Free and Go keep separate limits on uploads, image generation, voice, and data analysis, and ChatGPT tells you when you reach one. Go "may include ads," and the ways around them are in how to turn off ChatGPT ads.
How do ChatGPT's Codex and deep research limits work?
Codex and ChatGPT Work, OpenAI's agent for longer tasks, share one allowance with a five-hour window and a weekly one, much like Claude's. A new window starts "when you send your first message in Work or Codex after the previous window ends," and you need allowance left in both to keep working. Pro 100, Pro 200, and Pro 500 have no five-hour limit there, only the plan's overall allowance.
For Plus, OpenAI publishes how many local messages it expects a five-hour window to cover, which it calls "estimates, not guaranteed message counts."
| Model in Work and Codex | Plus, local messages per five hours (OpenAI's estimate) |
|---|---|
| GPT-6 Astra | 5 to 45 |
| GPT-6.1 Sol | 15 to 160 |
| GPT-6 Sol | 15 to 150 |
| GPT-6 Luna | 350 to 3,000 |
Cloud tasks can use more than local ones, and Fast mode spends the allowance at 2.5 times the standard rate. Past the allowance, Plus and Pro subscribers can buy credits for Codex, Work, Word, Excel, and PowerPoint, which expire 12 months after purchase. They can also buy an instant reset that refills both windows, though OpenAI says it "pulls your normal weekly allowance forward" rather than adding usage, and your next weekly period starts with your next request.
Deep research has a counter of its own. OpenAI says "Your in-product usage counter shows your remaining tasks," and on plans with a fixed monthly allowance it "resets every 30 days from the date of your first use." The pricing page calls deep research "Limited" on Free, "Expanded" on Plus, and "Maximum" on Pro, with no number for any of them.
How do I check how much usage I have left?
On Claude, open Settings > Usage. Its Plan usage limits section shows Current session, meaning the share of your five-hour limit used and the time remaining, and Weekly limits, with reset times for all models and for Fable where your plan includes it. Usage credits and any limit resets are listed on the same page.
On ChatGPT, Settings > Usage shows your Work and Codex allowance, its reset times, and your credit balance. In Chat, the limit message carries the reset time when ChatGPT has one, and deep research shows how many tasks remain. Don't count on support to restore a meter: OpenAI says its support team "does not reset ChatGPT or Codex usage limits," though it will investigate usage that looks miscounted.
How can I make Claude and ChatGPT limits last longer?
Each company publishes its own tips, and most of them cut how much the model has to read on every turn.
- Group related questions into one message. Anthropic suggests sending a batch, such as all your math problems at once, and reading your prompt over before sending so you need fewer follow-ups.
- Open a new chat for a new topic. Every turn sends the conversation so far, so a long chat costs more per message. Anthropic says chats long enough to trigger its automatic context management "consume more of your usage limit."
- Keep reference files in a Claude project. Anthropic says project content is cached and "counts less against your limits when reused," until the cache expires after a stretch of inactivity.
- Lower effort and switch off tools. Anthropic recommends a lower effort level, thinking off where the model has a toggle, and connectors and web search off when you don't need them. OpenAI says lower reasoning effort helps "make your allowance go further" in Work and Codex.
- Use a lighter model for routine work. Anthropic points to Haiku or the latest Sonnet ahead of Opus, and OpenAI suggests GPT-5.6 Luna for focused, repetitive tasks.
- Time heavy work to the window. Anthropic's credits guide says to "plan intensive work sessions around your five-hour usage windows," and OpenAI says to check Settings > Usage before starting a large task.
Can Claude and ChatGPT change their limits?
Yes, and both have this year. Anthropic says "We may adjust usage limits at our discretion," and its promotion that counted Fable 5 inside Pro's weekly limit ended on July 19, 2026. OpenAI says it canceled a cut to Pro 200's Work and Codex allowance that it had announced in a September 29 email, and doubled the lower allowance instead. Read any figure in a guide as a snapshot of the day it was checked.
How Morse keeps you chatting without watching a meter
Every limit above meters what one account sends, against windows each company sets and can change. Morse Pro is unlimited messages with no daily cap, for $19 a month or $180 a year, with Pro models such as Claude Sonnet 5.5, Grok 4.7, and DeepSeek V4.1 Flash Uncensored, and your choice of model in each chat. Pro also includes 250 monthly credits for the top models, such as Claude Fable 5.1 and GPT-6 Astra, and Max, at $49 a month, includes 5,500. Every plan is on the plans page, and the full breakdown is in Morse pricing.
Morse also changes who knows what you ask. You pick a privacy mode for each chat:
- Anonymous, to use Claude or GPT without the provider knowing it's you. The Morse Enclave sends your message on under Morse's name, so Anthropic or OpenAI never learns your name, email, account, or IP address. The provider still reads the message to answer it and may keep it for a limited time to check for abuse, so leave out names you'd rather keep to yourself. More in how to use Claude anonymously.
- Private, the default, for everyday questions you'd rather no host keep. Open-weight models like DeepSeek, Kimi, and GLM, plus xAI's Grok, answer under the zero data retention agreements Morse has with its inference, search, and other providers.
- E2EE, for anything no one else should read. Your browser checks the enclave's hardware attestation, then encrypts the message so only that enclave can decrypt it, not us and not the company running the enclave. E2EE chats come with Pro.
In every mode, your history is stored in your browser, encrypted, our servers keep no log of your messages, and no AI provider trains on what you send through Morse. Morse doesn't show ads. Anyone can check that the Morse Enclave runs the open-source code we published with npx @askmorse/verify inference.trymorse.ai, as Verify explains.
Start free with no card, and move to Pro when you want unlimited messages.
Sources
Read on November 20, 2026.
- Anthropic, Claude pricing
- Anthropic, What is the Pro plan?
- Anthropic, What is the Max plan?
- Anthropic, What is the Ultra plan?
- Anthropic, Getting started with Claude (Free plan limits)
- Anthropic, How do usage and length limits work?
- Anthropic, Usage limit best practices
- Anthropic, Manage usage credits for paid Claude plans
- Anthropic, What is a limit reset?
- Anthropic, Claude Fable models on your plan
- Anthropic, Claude Fable 5.2 launch offers
- Anthropic, Claude Cowork and chat are one Claude (how agentic tasks use your limit)
- OpenAI, ChatGPT pricing
- OpenAI, ChatGPT Free Tier FAQ
- OpenAI, What is ChatGPT Go?
- OpenAI, What is ChatGPT Plus?
- OpenAI, About ChatGPT Pro tiers
- OpenAI, GPT-6 and other models in ChatGPT
- OpenAI, Managing usage with GPT-6 Astra in Work and Codex
- OpenAI, Paid weekly Work and Codex rate limit resets
- OpenAI, Using credits for flexible usage in ChatGPT (Personal plans)
- OpenAI, Deep research in ChatGPT
Morse publishes this comparison. We've tried to keep it fair; if something's wrong or out of date, email hello@trymorse.ai.
Claude is a trademark of Anthropic. Morse is not affiliated with or endorsed by Anthropic.
ChatGPT is a trademark of OpenAI. Morse is not affiliated with or endorsed by OpenAI.
Common questions
Does Claude have a daily limit?
Not by that name. Anthropic's paid plans run on a five-hour session limit and a weekly limit, and its help pages add that it may also set "weekly and monthly caps" at its discretion.
How many messages do you get on Claude Pro?
Anthropic doesn't say. Its Pro page says the number "will vary based on message length," attached files, how long the conversation is, and the model or feature you use.
How long do I have to wait after hitting the Claude limit?
Until the five-hour session refills, counted from the first message of that session, or until your weekly reset if that's the limit you hit. Settings > Usage shows the time for both.
Do Claude Code and the Claude app share one usage limit?
Yes. Anthropic says usage across claude.ai, Claude Code, and Claude Desktop all counts toward the same limit.
Is ChatGPT Plus unlimited?
Everyday text chats are, on every ChatGPT plan, "subject to abuse guardrails." Thinking models, deep research, uploads, images, and Codex have their own limits, and OpenAI's Plus page says caps can tighten "especially during high demand."
What happens when I hit a limit in ChatGPT?
OpenAI says ChatGPT "may continue with another available model" when you reach a Thinking limit. Otherwise you wait for the reset time it shows, or, for Codex and ChatGPT Work on Plus and Pro, buy credits.
Does upgrading reset my usage limit?
On Claude, Anthropic says a new plan's limits "apply right away." On ChatGPT, OpenAI's Free tier FAQ says your usage rates reset when you move from Free to Plus, Pro, or Business.
Can support reset my ChatGPT or Claude limits?
OpenAI says its support team does not reset ChatGPT or Codex limits, though it will look into usage that seems miscounted. Anthropic gives eligible plans occasional limit resets, which appear under Limit resets in Settings > Usage.
Do Claude and ChatGPT usage credits expire?
Anthropic's usage credits bought on or after December 7, 2026 expire 12 months after purchase, and earlier ones don't. OpenAI's credits expire 12 months from purchase, or 180 days for purchases in Japan after November 13, 2026.
Does Morse Pro have a daily cap?
No. Morse Pro is unlimited messages with no daily cap, at $19 a month or $180 a year.
Does the AI provider know it's me when I use Claude or GPT through Morse?
No. Every message goes through the Morse Enclave, which sends it on under Morse's name, so the provider never learns your name, email, account, or IP address. In Anonymous mode the provider still reads the message to answer it, and may keep it for a limited time to check for abuse.