How Many Tokens Are Included in Claude Pro? Limits Explained
Claude Pro has no fixed token allowance. Learn how its model context windows differ from session and weekly usage limits, and how to check your own usage.
- Anthropic does not publish one fixed token allowance for every Claude Pro subscriber.
- Context windows describe one conversation, while session and weekly limits control usage over time.
- Some current paid models support up to one million tokens, but other models and product surfaces have smaller windows.
- Settings, then Usage shows the session and weekly limit information for your account.
Claude Pro does not come with one fixed number of tokens that every subscriber can spend. Anthropic separates two limits: the context window, which controls how much information a model can work with in one conversation, and usage limits, which control how much you can use Claude over time. A selected paid model can support a very large context window while the account still has a session or weekly usage limit.
Short Answer
There is no published fixed token allowance for Claude Pro. Anthropic says Pro usage depends on message length, attached files, conversation length, the model and the features you use. Paid plans have a five-hour session limit and a weekly limit. You can see the current session and reset information in Settings, then Usage.
The context window is a different number. Anthropic's current paid-plan table lists some newer Claude models with a one-million-token context window, other models at 500K and some at 200K. That number describes how much information can fit in one conversation on a supported model. It is not a monthly token balance or a promise of unlimited messages.
What the Question Really Means
People often use "token limit" to describe three different things. First is the context window: the maximum amount of text and other supported input that the model can consider at once. Second is the plan usage limit: how much work your account can do during a time period. Third is the output limit: how much the model can generate in one response.
Claude Pro's subscription mainly changes the amount of usage you receive compared with the Free plan. It does not give every subscriber a simple token wallet such as a fixed number of tokens per month. Anthropic describes a five-hour session limit and a weekly limit, and says usage varies with the work you send.
This is why two Pro users can reach a limit after different amounts of work. A short question with a short answer uses less of the allowance than a long conversation with large files, a reasoning-heavy model and several tools. The model you select also affects how much usage a task consumes.
How It Works
Think of the context window as a desk. It is the amount of information Claude can keep available while working on one task. The usage limit is more like the amount of time or work available to your account during a period. A large desk does not mean you can work without a break.
Anthropic's current context-window page lists one-million-token windows for selected paid chat models, including some of its latest Fable, Opus, Sonnet and Haiku models. Other models in the table have 500K or 200K windows. The table also separates Claude chat, Claude Code and Claude Cowork. The values can differ by model and product surface.
Do not read "one million tokens" as a guarantee that every Pro conversation can hold one million tokens. The model must support that context size, the relevant product surface must expose it, and some special model variants or usage-credit requirements can apply. The context window also includes the space needed for Claude's response and system instructions, so the full number is not all available for user input.
How Claude Pro Usage Limits Work
Anthropic says paid plans have a five-hour session limit and a weekly usage limit. The five-hour period starts when you send your first message. If you use the available allowance before the period ends, you wait until the reset. The weekly limit applies across models, and some plans can have an additional model-specific weekly limit.
Your usage is affected by the length and complexity of the conversation, attached files, the model, selected effort level and the features you use. A long research task or code workflow can consume more than a short text exchange. Tool use also matters because search, connectors, code execution and other features add work to a session.
The limits are shared across Claude's product surfaces. Anthropic says activity in claude.ai, Claude Code and Claude Desktop counts toward the same account usage limit. Moving from the web app to the terminal does not create a separate Pro allowance.
Anthropic currently lists Pro at $20 per month in the United States, with local currency and tax treatment varying by region. That price does not buy a fixed token bucket. Pro gives more usage per session than Free, priority access during busy periods and access to Claude Code, but the usage rules still apply.
Practical Steps
Open Claude and go to Settings, then Usage. The Current session area shows how much of the five-hour allowance you have used and the time remaining. The weekly section shows the reset information assigned to your account. The exact labels can change as Anthropic updates its settings pages.
Check this page before a long task. If you are close to a limit, save your work and decide whether to finish a smaller part now or wait for the reset. Do not assume that starting a new chat will reset a weekly limit. It only gives you a new conversation context; the account usage period still applies.
If you use Claude Code, keep an eye on the same account usage. A coding session can use the same Pro allowance as a long conversation in the web app. For repeated code work, split tasks into small steps and keep only the necessary project files and tools active.
How to Use Your Context Window More Efficiently
Start with a short task brief. Tell Claude the goal, the files it should use and the result you need. Avoid pasting a large history of unrelated messages when a compact summary will do. Keep project instructions focused on stable facts and put task-specific directions in the current prompt.
Use Projects when you work with a large collection of reference files. Anthropic says Projects use retrieval-augmented generation to load relevant material instead of adding every file to the context window at once. Remove files that no longer matter and keep instructions short. This can help both the context window and usage limits.
Choose the model and effort level based on the task. A routine summary may not need the most capable model or the highest effort. If the model offers a thinking setting, use it when the task benefits from deeper reasoning and turn it down for routine work. Anthropic notes that tools and connectors are token-intensive, so disable ones you do not need.
For a long code or research task, ask Claude to work in stages. Request a plan, confirm it, then ask for one stage at a time. This makes it easier to review results and reduces the chance that one oversized request consumes a large part of the available allowance.
Common Problems
A long conversation can use more context and more usage than a short one. Large attachments add more material for Claude to process. Research and coding tasks can involve several model steps, tools and intermediate outputs. A high effort setting can also use more tokens than a quick response.
Automatic context management can help with long conversations when code execution is enabled. Anthropic says Claude can summarize earlier messages as a conversation approaches the context window limit. This allows the conversation to continue, but the longer session can still consume more of the plan's usage allowance.
If you hit a limit, check whether it is a session limit or a weekly limit. A session limit has a five-hour reset. A weekly limit resets at the time assigned to your account. Anthropic also offers usage credits for eligible paid plans, but you must enable them and review the billing terms before using them.
What Does Not Work
Do not assume that every Claude model has a one-million-token context window. The current table shows different values by model and product surface. Do not assume that a one-million-token context window gives you one million tokens to spend every week. It is a capacity limit for one conversation, not a subscription allowance.
Do not assume Pro includes Claude API usage. Anthropic says Pro does not include API usage through the Claude Console. If you are building an app with the API, you need separate Console access and billing. Also do not assume that Max or other plans remove every usage limit. Higher plans offer more capacity, but Anthropic still documents limits.
Safety and Limitations
The context-window table and usage rules can change as Anthropic adds models and updates plans. Check the official support pages before making a purchase decision or planning a large project. Avoid unofficial calculators that claim to know your exact monthly Pro token allowance, because Anthropic does not publish a fixed number for every subscriber.
For important work, save outputs outside Claude and keep a record of source files and decisions. A limit notice should not be the only copy of your work. If your account shows a different limit from a help page, trust the in-product Usage page for your account and contact support if the difference looks like an error.
Key Takeaways
- Claude Pro does not have a fixed public token allowance.
- Context windows and plan usage limits are different things.
- Some current paid models support up to one million tokens, while other models have smaller windows.
- Pro has five-hour and weekly usage limits shared across Claude's product surfaces.
- Settings, then Usage is the best place to check your account's current limits.
Related Reading
For plan details, see Is Claude Pro worth it? and How much is Claude?. For model selection, read What Claude model is best for coding?.
FAQ
How many tokens does Claude Pro include?
Anthropic does not publish a fixed token allowance for Pro. The plan uses session and weekly usage limits, and the amount of work you can complete depends on message length, files, model and features.
Does Claude Pro have a one-million-token context window?
Some paid models have a one-million-token context window. Other models and surfaces can have 500K or 200K windows. Check Anthropic's current table for the model you plan to use.
Is context the same as usage?
No. Context is the amount of information Claude can work with in one conversation. Usage is the amount of work your account can do during a session or week.
When does the Claude Pro limit reset?
The five-hour session period starts with your first message and resets five hours later. The weekly limit resets at a fixed time assigned to your account. Check Settings, then Usage.
Does Claude Code count toward my Pro limit?
Yes. Anthropic says usage across claude.ai, Claude Code and Claude Desktop counts toward the same account usage limit.
Does Claude Pro include API tokens?
No. Pro is for the Claude chat experience and does not include API usage through the Claude Console. API usage is billed separately.