The short answer
When you sign in to Claude Code with a Pro, Max or Ultra subscription, it draws from the same limits as claude.ai (Anthropic's help center). That means a weekly limit and, except on verified Ultra accounts, a session limit that resets every five hours, plus on Max and Ultra a separate weekly limit for Fable. Chat and coding share one pool, so a long Claude Code session can leave you blocked on claude.ai too. With an API key instead, the subscription limits do not apply. An ANTHROPIC_API_KEY environment variable makes it bill the API rather than your subscription (per the same article).
Which limits apply to you
| How you use Claude Code | Limits | Where to see them |
|---|---|---|
| Pro, Max or Ultra subscription | Shared with claude.ai: the weekly limit, the 5-hour session except on verified Ultra accounts, and on Max and Ultra a separate weekly limit for Fable (help center, Ultra) | /usage inside Claude Code, claude.ai/settings/usage, or Claude Monitor in your browser |
| API key (Claude Console) | No subscription limits; the key is billed as API usage (help center) | The Claude Console, or the Session block at the top of /usage inside Claude Code, an estimate of the session's API cost (Claude Code docs) |
For how each limit works, see Claude usage limits explained; for the reset times, when does the Claude usage limit reset.
Why Claude Code uses your limit faster than chat
Anthropic's help center lists conversation length, attachments, tools, multi-step tasks, model and effort among what affects usage (how usage limits work, Usage limit best practices). Coding hits several at once.
- Agentic loops. One request can turn into many model calls as Claude Code reads files, runs commands and edits code.
- Large context. Each call carries the conversation so far plus the files it has read, so long sessions cost more per step (Claude Code docs).
- Stronger models. The most capable model uses more of your limit per message than a lighter one.
How to make your limit last
- Clear or compact between tasks. Use
/clearwhen you switch to an unrelated task and/compactto shrink a long conversation, so each call carries less context. - Match the model to the job. Switch with
/modeland keep the strongest model for the hard parts. - Plan around the reset. If the session window is nearly used up and resets soon, wait for the reset before starting a big refactor. Anthropic is gradually introducing a wrap-up allowance that lets Claude Code finish the step in progress after you hit the session limit, once per weekly period on Pro and each session on Max; it counts toward the weekly limit (help center).
- Keep the numbers in view. Running
/usageor checking the usage page by hand is easy to forget mid-task. A tracker in the toolbar shows the session and weekly percentages and warns you at 80% and 95%.
Tools that help
Claude Monitor (ours, free and open source) reads the limits Claude reports and shows the session, weekly limit, per-model caps, usage credits and routine runs in your browser toolbar, with a countdown to each reset. Because the limits are shared, the numbers you see there include your Claude Code usage.
Claude Code's own /usage command shows the same plan usage bars, a usage-credits row when credits are enabled, and falls back to the last known usage when it cannot fetch fresh numbers (Claude Code docs). With an API key, the Session block at the top of /usage shows an estimate of the session's API cost instead. For what to look for in a tracker, see choosing a Claude usage tracker.
With the browser closed: UsagePeek AI
UsagePeek shows your usage next to the clock with the browser closed, and reopens your Claude Code projects the moment a limit resets.
Per its Microsoft Store listing, it shows usage next to the clock and resumes Claude Code in the Claude app or terminal on reset. Auto-resume requires the Claude Code CLI. Windows 10 and 11.
Free for 7 days, then $9.99 one-time, no subscription (Microsoft Store listing, October 18, 2026). Same maker as this extension, separate app and purchase; the extension stays free.
Questions
Does Claude Code share usage limits with claude.ai?
Yes, on Pro, Max and Ultra plans. Claude Code and chat on claude.ai draw from the same limits, so usage in one reduces what is left in the other (Anthropic's help center).
How do I check my Claude Code usage limit?
Inside Claude Code, run /usage: it shows your plan usage bars, plus a usage-credits row when credits are enabled; when it cannot fetch fresh numbers it falls back to the last known usage (Claude Code docs). The claude.ai usage page and Claude Monitor show the same limits.
When does the Claude Code limit reset?
The session limit resets every five hours, and the weekly limit once a week at a fixed time assigned to your account (help center). Because the limits are shared with claude.ai, the reset times are the same ones claude.ai shows in Settings > Usage, and the extension counts down to each of them. More in when does the Claude usage limit reset.
What happens when Claude Code hits the limit?
If the session limit arrives mid-response, Claude Code may finish the step in progress; this wrap-up allowance is rolling out gradually (help center). After that, you wait for the reset. With usage credits enabled, you keep working at standard API rates up to a monthly spend limit you set (help center). The resets do not change.
Does using an API key avoid the limits?
An ANTHROPIC_API_KEY environment variable makes Claude Code use the key instead of your subscription (help center). You then pay API usage charges instead of drawing on the subscription's included usage. It is a different cost model, not a bigger allowance.
Do Claude Code routines count against my limits?
Anthropic's help center does not document a routines limit. As of October 2026, the usage data the extension reads includes a daily run count for routines. Claude Monitor shows it as a used and limit count.
Keep your Claude Code limits in view
The extension shows the shared session and weekly limits in your browser toolbar; UsagePeek AI shows them by the Windows clock.