Claude Code Usage Limits: Pro, Max, and API (2026)
Claude Code (Anthropic’s command-line coding agent) doesn’t have its own separate quota — it draws down the usage on whatever plan you sign in with. On a Claude Pro or Max subscription it shares the same 5-hour rolling window as Claude chat, plus a weekly cap layered on top; Max 5x and Max 20x raise both ceilings. If you’d rather not deal with subscription windows at all, you can run Claude Code on API credits (pay-as-you-go), where you’re billed per token instead of throttled by a window. These caps change often — treat the numbers as a snapshot.
The Limits at a Glance (as of 2026)
Pro ($20/mo)
- Shares Claude’s 5-hour rolling window plus a weekly cap
- Fine for light or occasional coding sessions; heavy agentic runs exhaust it fast
Max 5x ($100/mo)
- Roughly 5x Pro’s usage in both the 5-hour window and the weekly cap
Max 20x ($200/mo)
- Roughly 20x Pro’s usage — the highest subscription tier for sustained coding
API (pay-as-you-go)
- No fixed usage window — you’re billed per input/output token
- Governed by account rate limits (requests + tokens per minute), not a 5-hour cap
Anthropic tunes these caps frequently and Claude Code’s heavy tool use burns through a window faster than chatting. Check Anthropic’s help center or run
/statusin Claude Code for your current usage.
Why Claude Code Eats a Window Faster Than Chat
Claude Code is an agent, not a chat box — a single instruction can trigger dozens of file reads, edits, and tool calls, each consuming tokens against your allowance. That means the same 5-hour window that lasts all day when you’re chatting can be drained in a couple of hours of intense agentic coding. Long context (large files, whole repos) draws it down faster still.
Two Caps at Once: 5-Hour Window + Weekly Limit
On Pro and Max, Claude Code shares Claude’s dual-layer system. The 5-hour rolling window starts with your first request and refreshes 5 hours later. On top sits a weekly cap that resets 7 days from your first message — added so that continuous heavy use across many sessions still has a ceiling. You can hit the weekly cap even if no single 5-hour window is maxed out.
Subscription vs API: Which Cap Model Fits You
The subscription path (Pro/Max) gives you predictable monthly cost but a hard window — great if your usage is bursty and you can wait out a reset. The API path removes the window entirely: you pay per token and are limited only by account rate limits, so an all-day heavy run never gets throttled by a 5-hour cap, but the bill scales with usage. Many heavy users keep a Max subscription for daily work and fall back to API credits when they need to blow past a weekly cap.
How to Avoid Hitting the Limit
- Scope your prompts — point Claude Code at the specific files or directory it needs instead of the whole repo.
- Use
/statusto watch how much of the window you’ve spent. - Save big agentic runs for when your window has room, or run them on API credits.
- Upgrade Pro to Max 5x/20x if you code heavily every day, or switch to the API for uncapped throughput.
Troubleshooting
Does Claude Code have its own usage limit?
No — it consumes the usage on the plan you log in with. On Pro/Max it shares Claude’s 5-hour window and weekly cap; on the API it’s billed per token.
How long until my Claude Code limit resets?
The 5-hour window resets 5 hours after your first request. The separate weekly cap resets 7 days from your first message.
Can I use Claude Code without a subscription?
Yes. Authenticate with an Anthropic API key and you pay per token with no 5-hour window — only standard account rate limits apply.
Why does Claude Code hit the limit faster than Claude chat?
Because it’s an agent: one instruction can fire many file reads, edits, and tool calls, each of which counts against your allowance.
Does Max really give more Claude Code usage?
Yes. Max 5x is roughly 5x Pro and Max 20x roughly 20x Pro in both the 5-hour window and the weekly cap.
Quick Reference
| Plan | Cap model | Relative headroom |
|---|---|---|
| Pro ($20/mo) | 5-hour window + weekly cap | Baseline |
| Max 5x ($100/mo) | 5-hour window + weekly cap | ~5x Pro |
| Max 20x ($200/mo) | 5-hour window + weekly cap | ~20x Pro |
| API (pay-as-you-go) | Per-token billing + rate limits | No fixed window |
Figures as of 2026 and subject to frequent change — check Anthropic’s docs for current caps.
If you’d rather have an AI that does work in your inbox and calendar than one you run in a terminal, an assistant like Carly runs from your email and acts on triggers — see the best AI personal assistants.
Related guides: Claude usage limit · Cursor usage limits · GitHub Copilot limits · Claude file upload limit · DeepSeek limits
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
