If you've ever seen "you've reached your usage limit" in Claude, here's the key idea: Claude's usage limit isn't a simple "N messages per day" — it's based on how much you use within a set time window. This guide explains, based on Anthropic's official Help Center, how limits are set, when they reset, and how to check and reduce your usage.
🟢 Model references match the current lineup · model notice · Fable subscription
Fable 5 and 5.1 subscription (updated September 7, 2026): Claude Fable 5.1, released September 1, 2026, is the current Fable model and Fable 5 is now legacy. Plan terms are the same for both — Max and Team Premium plans include Fable at up to 50% of the weekly usage limit; Pro and Team Standard use usage credits (
The key: it's "usage over time," not "message count"
Claude doesn't just count messages. Officially, how many messages you can send depends on message length, the length of attached files, your current conversation length, the model or feature you use, the effort level you've selected, and current demand. Usage across all Claude surfaces — claude.ai, Claude Code, and Claude Desktop — counts toward the same limit. A message sent at the end of a long conversation consumes far more than one in a short new chat. That's why Anthropic does not publish a fixed message count like "X per day."
When does it reset? (Every 5 hours)
For both Free and Pro, the session-based usage limit resets every five hours — so even if you hit a limit, you can use Claude again after the window passes. Paid plans (Pro and Max) additionally have a weekly usage limit that, per official guidance, resets at a fixed time each week assigned to your account (the reset day and time stay the same regardless of when you start). In particular, Max plans have two weekly limits — one across all models and one for a specific model group (official). The model-group name shown on your screen can vary by account and period, so check Settings > Usage for the exact split and reset times.
September 22, 2026: higher five-hour limits plus a reset you can save
In the Claude Opus 5.5 announcement (September 22, 2026) Anthropic said it is increasing five-hour usage limits on Pro, Max, Team and seat-based Enterprise plans and giving subscribers a rate limit reset they can save and use whenever they choose. The size of the increase and the number, expiry and mechanics of resets are not stated in official materials (checked September 27, 2026). There is no official "doubled usage" figure; the only confirmed number is the 25% raise to Claude Code weekly limits from September 14. Check Settings > Usage in the app to see whether a reset is available to you.
How do I work out when my limit resets?
A five-hour session starts when you send the first message of that session and resets five hours later. If your first message went out at 9:00, that session's limit clears at 14:00. When you hit the limit the screen shows the time remaining, and Settings > Usage shows the next reset time (Pro and above). The reset time depends only on the start time, not on how much you used, so if you have urgent work it helps to note when a session began. Weekly limits reset separately at a fixed time assigned to your account.
Claude Code weekly limits — 50% higher now, a permanent 25% raise from September 14, 2026
Claude Code weekly usage limits are currently running 50% higher than normal. This covers Pro, Max, Team, and seat-based Enterprise plans, and it applies automatically with nothing to redeem. Anthropic's developer account (@ClaudeDevs) announced it; the window originally ran through July 13 and was extended three times — to July 19, August 19, and August 31.
On August 30, 2026 the direction changed. Anthropic said it would permanently raise the standard weekly limit by 25% starting September 14, and the official help article now lists the promotion as running May 13 – September 13, 2026 (checked September 14, 2026), confirming the switch. So this is not the promotion lapsing and limits reverting — the standard limit itself is being reset to a level 25% above the old baseline.
Measured against what you have today, however, the limit goes down. Anthropic said so directly in the same announcement, stating that compared to today this works out to a 17% reduction. Normalizing the old baseline to 100, today is 150 and from September 14 it becomes 125 (an illustration of the ratio only — absolute figures are not published).
Two things to keep separate. First, this applies to weekly limits only — the five-hour session limit described above is unchanged, and the August 30 announcement made no mention of it either. Second, nothing about your plan or billing changes across the transition, and no action is needed on your part. If you have limit-heavy work, the window before September 14 is the roomier one. For the full announcement and background, see the September 14 change.
- The limit is based on usage within a time window (about every 5 hours), not a message count.
- Pro gives roughly 5x the free usage (per official guidance; actual amount varies with demand and conversation length).
- Hitting the limit isn't permanent — you can use it again after the reset. Long chats, large files, and higher-end models burn through usage faster.
- Exact free-tier numbers are not officially published and adjust dynamically with demand.
Free vs Pro: how big is the gap?
Officially, the Pro plan offers at least five times the per-session usage of the free service during peak hours. The reset cycle (5 hours) is the same; what differs is how much you can do before hitting the wall. As an official illustration, with relatively short conversations (about 200 English sentences of 15-20 words) on a lighter model, you can expect around 45 messages every five hours — often more, depending on capacity. This is a conditional example; actual numbers vary by conversation length, model, and demand.
How to check your usage
Claude shows a notification when you reach a limit. On paid plans (Pro, Max, Team, etc.), open Settings > Usage (or go directly to claude.ai/settings/usage) to see progress bars for (1) your current 5-hour session usage and time remaining, and (2) when your weekly limits reset (all models and a specific model group). The free plan has no usage dashboard — you only get an on-screen warning as you near the limit. Pro and Max users can also enable usage credits to keep working at standard API rates after exhausting the included limits.
📍 Exact location (official, 2026): Per Anthropic's usage limit best practices, paid plans show session and weekly progress bars in the "Plan usage limits" section at the top of Settings > Usage.
How to reduce your usage (officially recommended)
From Anthropic's official "usage limit best practices":
- Start a new conversation instead of continuing a very long one. Claude re-reads the whole conversation each turn, so longer chats cost more per message.
- Batch your questions into one message, especially with long documents.
- Don't re-upload the same file within a conversation — Claude remembers it.
- Put frequently referenced material in Projects — the more you reuse content, the more you benefit from caching.
- Turn off extended thinking when unneeded, and enable token-heavy tools (web search, Research, MCP connectors) only when needed.
- Pick a lighter model suited to the task — bigger models cost more per message.
"Usage limit" is not "length limit"
These are easy to confuse. Usage limit = how much you use overall within a time window; length limit = how long a single conversation can get (the context window). Officially, the chat context window varies by model — on paid plans, Fable 5.1, Opus 5, and Sonnet 5 support 1M tokens, Opus 4.8/4.7/4.6 and Sonnet 4.6 support 500K tokens, and other models 200K. Claude Code differs slightly: on Pro, Max, Team and Enterprise plans, Fable 5.1, Sonnet 5, Fable 5, Opus 5, Opus 4.8, 4.7 and 4.6 all support 1M tokens, though Pro users must enable usage credits to get the 1M window on Opus models. When code execution is enabled, Claude now automatically summarizes earlier messages as a conversation nears the context limit so it can continue seamlessly (your full history is preserved) — you may see a "Claude is organizing its thoughts" note while this happens.
Note (checked September 7, 2026): the plan comparison table on the official pricing page (claude.com/pricing) lists the context window as 200K for individual plans and 500K on the default model for Enterprise, which differs from the support article this guide follows (1M for Opus 5 and Sonnet 5). The limit actually applied can depend on plan, model, and settings, so when in doubt check your settings screen and both official pages.
Frequently asked questions
How many times a day can I use Claude on the free plan?
There is no officially published fixed "N per day." The limit works as a roughly 5-hour usage window and shifts dynamically with current demand, conversation length, and the model you use, so an exact count can't be stated.
When does the usage limit reset?
Per official guidance, it resets about every 5 hours. When you hit the limit you can wait for the next reset, switch to a lighter model, or upgrade your plan.
What happens when I hit the limit?
Further use is restricted for that usage window. Once it resets, you can use Claude again. Your past chats and projects are not deleted.
If I use up just one model's weekly limit, am I fully blocked?
On Max plans the weekly limits are tracked separately — one across all models and one for a specific model group (official). So exhausting the model-group limit alone doesn't necessarily block everything: if your all-models limit still has room, you can likely keep working with other (usually lighter) models, while the exhausted group stays restricted until its weekly reset. If you exhaust the all-models limit, usage is broadly restricted until reset. To keep going you can (1) wait for the reset, (2) upgrade your plan, or (3) enable usage credits to continue at standard API rates (official). Note: the exact behavior when a limit is hit (hard block vs. fallback to another model) and the model-group label shown can vary by account, period, and policy — confirm the current state in Settings > Usage and the official docs (this part may not be certain).
How much more does Pro give me versus free?
Per official guidance, roughly 5x the free usage. This is an approximate baseline that also varies with demand and how you use it.
What burns through usage the fastest?
Long conversations (more accumulated context), large file/image attachments, and higher-end models (such as Opus) consume usage faster. Splitting work into shorter, focused chats helps conserve it.
Next steps
For free vs paid differences, see the free plan explainer; for plan pricing, see the plan comparison. If a limit screen looks like an error or you also can't connect, see what to do when Claude isn't working.
Related: Billing and subscription guide · Free plan guide · API key billing guide
Related: Claude Code Pricing Explained: Plans, API Costs & Limits
For other billing, subscription, and usage issues (cancellations and refunds, card changes, limits, receipts), see the Claude billing, subscription and usage hub.
Related articles
- Claude Free Plan Explained: Limits, Usage, and Free vs Pro
- Manage Claude Subscription & Billing — Cancel, Refund (2026)
- How to Reduce Claude Usage: 6 Habits to Hit Limits Less
- Claude Code Pricing Explained (2026): Plans, API Costs & Limits
- Claude Pricing 2026: Free vs Pro vs Max — Which to Pick?
- How to Upgrade Your Claude Plan — Free to Pro to Max