Claude Code introduces the Wrap-Up Allowance: If Claude Code reaches your plan’s 5-hour usage limit partway through a response, it keeps working briefly so it can reach a reasonable stopping point rather than stopping mid-step. While happening, the status displays “Usage limit reached · wrapping up”. Usage during wrap-up counts toward your weekly limit. Available on Pro (up to once per weekly usage period) and Max / Team Premium seats (each time a 5-hour limit is reached).
https://support.claude.com/en/articles/17040437/anthropic — Claude Code usage limit resets
LAST RESET
2026-10-14
A change to the limit rules themselves, affecting your allowance going forward.
24
Total records
1
Banked resets
13.0d
Avg. interval
47.6d
Longest wait
SINCE LAST RESET35d
Average gap 13d · longest 48d · 90% of past waits were shorter than this one
$ Anthropic posts resets and limit changes on @ClaudeDevs — both outright fleet-wide resets and changes to the rules themselves. Both kinds are tracked here.
// ACTIVITY · 52 WEEKS UTC
regular banked policy
// LOG · 24 entries
newest firstIf you're on Pro, Max, or Team, your reset is available today in Settings → Usage. Apply it any time until Oct 22. Opus 5.5 is the default for paid plans. It's priced lower than Opus 5, so your 5-hour and weekly limits go 25% further. — second tweet of the Opus 5.5 launch thread (x.com/ClaudeDevs/status/2102438800836489554), which opens with: Opus 5.5 performs at the level of Fable 5.1. It's ~30% faster and ~40% cheaper than Opus 5 per task. In Claude Code: - 5-hour session limits increase 20% today - Opus 5.5 is priced lower, so it goes 25% further within limits - Pro, Max, and Team users get a reset to use anytime
https://x.com/ClaudeDevs/status/2102438803013333469Alongside the Fable 5.1 release, Anthropic reset the 5-hour and weekly limits for all users.
https://x.com/ClaudeDevs/status/2094856679250919746Fable 5.1 keeps the same pricing as Fable 5, but cache-read pricing drops from $1/MTok to $0.25/MTok. For long-horizon workloads, pairing this with prompt caching saves a lot.
https://x.com/ClaudeDevs/status/2094851235694497836Official announcement: from 14 September, Claude Code's standard weekly limit rises permanently by 25% for Pro, Max, Team and seat-based Enterprise; until then the current 50% boost stays in effect. Anthropic also noted that, relative to the promotional level, this amounts to roughly a 17% reduction in the weekly limit.
https://x.com/ClaudeDevs/status/2093742321473065266The 50% boost to weekly limits is extended through 31 August. Anthropic says it hopes to make the increase permanent, but strong demand for its models means capacity may be tight in the coming weeks.
https://x.com/ClaudeDevs/status/2089798442306711646Claude Code desktop adds an "auto-continue" option: turn it on after hitting the cap and it picks up where it left off as soon as the limit resets.
https://x.com/ClaudeDevs/status/2088014831605702937Auto mode runs a classifier on every tool call. As of today, those extra tokens no longer count against usage limits on Pro, Max and Team. Enterprise and API users get the same treatment once auto mode becomes their default too.
https://x.com/ClaudeDevs/status/2085794868102889473// NOTIFY ME
Email me when Claude Code resets
One email per official reset. No digest, no marketing, one-click unsubscribe. Prefer X? Reset alerts are also posted at @XIAOMIAO_26 · RSS
// PLANS
| Plan | Price | 5h gate |
|---|---|---|
| Free $0 Active Enough to try it out; no Claude Code | ||
| Pro $20/mo Active At least 5× the free tier per session | ||
| Max 5x $100/mo Active About 5× Pro per session; Fable capped at 50% of the shared weekly allowance | ||
| Max 20x $200/mo Active About 20× Pro per session; Fable capped at 50% of the shared weekly allowance | ||
| Team / Enterprise per seat Active Allocated per seat; also got the 5-hour doubling from 2026-05 | ||
Only relative multiples are published, never absolute token counts; usage is metered on input + output tokens combined.
// WHERE TO CHECK
Claude Code
Type /usage to see both windows; /status also shows your current allocation
Web settings
Settings → Usage shows your weekly reset time, the overall allowance and, on Max, the Fable cap within it
The limit message
When you are rate limited, the message itself states the reset time — go by that
// ERRORS
- “5-hour limit reached ∙ resets 5pm” Claude Code
- “Weekly limit reached ∙ resets Thu 9am” Claude Code
- “Rate limited, please try again later” Claude Code
- “You're out of messages until 5 PM” Claude (web & desktop)
// FAQ
01 When does the Claude Code weekly allowance reset?
At a fixed time each week, assigned to your account and unrelated to when you start using it. You can find your own reset time under Settings → Usage on the web. This is the biggest mechanical difference from OpenAI, whose weekly window rolls from your first request.
02 How does the 5-hour session work?
It opens when you send your first message and resets 5 hours later. It also drains on a rolling basis: usage from roughly 5 hours ago frees up over time, so bursts are fine — only sustained hours of full-throttle agent work will empty it.
03 Does Fable have its own weekly limit?
Max has a general weekly limit plus a Fable weekly cap. Fable uses the same overall allowance, with at most 50% available for Fable; it is not an extra pool. Claude and Claude Code share these limits.
04 What changed on 14 September 2026?
Since 13 May 2026 Claude Code weekly limits had been running at 150% of standard as a promotion. That promotion ended on 14 September, replaced by a permanent 25% raise over the pre-July baseline. Relative to the promotional level, that is a net cut of about 17%.
05 Are Claude and Claude Code allowances separate?
No. On Pro and Max plans the allowance is shared between the Claude web/desktop apps and Claude Code — activity on both sides draws from the same pool.
06 How do I check what I have left?
Type /usage in Claude Code to see both windows; /status also shows your current allocation. When you are rate limited, the message states the reset time directly — go by that.
07 What does "5-hour limit reached ∙ resets 5pm" mean?
Your 5-hour session is exhausted. The message shows its reset time in your local time zone. Switching models does not refresh this allowance. Check /usage for the separate weekly limits; wait for reset or use usage credits.
08 What does "rate limited, please try again later" mean in Claude Code?
Either your 5-hour session or your weekly allowance is exhausted — the message usually names which and gives the reset time. If neither bar is full, it is a temporary capacity throttle on Anthropic's side; those clear in minutes and Claude Code retries on its own. Check status.anthropic.com if it persists.
09 How do Pro, Max 5x and Max 20x limits compare?
Pro costs $20/month, Max 5x $100/month and Max 20x $200/month. Max provides approximately 5× or 20× Pro usage. All retain a 5-hour and general weekly limit; Max also caps Fable within that weekly allowance.
10 How can I get more Claude Code usage?
Three ways: wait for the window to reset, buy usage credits (formerly "extra usage") so work continues past the cap at pay-as-you-go rates, or upgrade to a Max tier. Anthropic also occasionally resets everyone — this page logs each one. There is no legitimate "hack"; anything claiming to bypass limits violates the terms.
11 Does "reset" mean my usage or my conversation?
On this site, always usage. Inside Claude Code, /clear resets the conversation context and /compact condenses it — neither touches your allowance. If you searched "claude code reset" looking to clear memory or settings, that is a different thing from the limit resets tracked here.
12 Do thinking tokens, subagents, prompt caching and web search count toward the Claude Code limit?
All of them draw from the same allowance. Extended thinking is billed as output tokens. Each subagent is its own model call with its own context, so fanning out five subagents costs roughly five conversations. Cache reads count at a reduced rate rather than being free. Web search results are injected as input tokens on top of the search itself. Anthropic publishes no exact multipliers for subscription plans — /usage after a session is the only reliable way to see what a workflow costs.
13 What time does Claude Code reset in my time zone?
The limit message and Settings → Usage already show your 5-hour and weekly reset times in local time — no conversion needed. There is no global clock time: the weekly reset is a fixed slot assigned per account, so two people in the same city can reset on different days. Fleet-wide resets announced on @ClaudeDevs apply immediately, whatever the hour.