Guide
Claude Code usage limits by plan, and what to do when you hit them
Anthropic publishes multiples, not message counts. What each plan gets, how to check what's left, and the ways past the limit.
Claude Code has no usage limit of its own: it draws on your Claude plan's allowance, shared with Claude on the web, desktop and phone. Every paid plan has a session limit that resets every five hours and a weekly limit across all models. Anthropic publishes no message counts, only multiples of Pro: Max 5x gets five times Pro's per-session allowance, Max 20x twenty times, Team standard seats 1.25 times and premium seats 6.25 times. Enterprise's current seat has no per-seat limit and bills usage at API rates.
Key takeaways
- Chat and Claude Code count against the same limits.
- Two limits: a five-hour session limit and a weekly limit that resets at a fixed time assigned to your account.
- Team limits are per member, not pooled across the team.
- Past the limit: wait, turn on usage credits (formerly extra usage) at API rates, use a saved limit reset, switch to an API key, or upgrade.
- In 2026 Anthropic doubled five-hour limits on May 6, raised them again on September 22, and set weekly limits 25% above standard from September 14.
Claude Code usage limits by plan
| Plan | Price | Five-hour allowance | Weekly limit |
|---|---|---|---|
| Pro | $17 a month billed annually, $20 monthly | The baseline | Yes, across all models |
| Max 5x | $100 a month | 5 times Pro | Yes |
| Max 20x | $200 a month | 20 times Pro | Yes |
| Team, standard seat | $20 a seat billed annually, $25 monthly | 1.25 times Pro, per member | Yes |
| Team, premium seat | $100 a seat billed annually, $125 monthly | 6.25 times Pro, per member | Yes |
| Enterprise (current seat) | $20 a seat plus usage at API rates | No per-seat limit | No; billed per token |
| API key (Claude Console) | Per token | No plan limit | No; billed per token |
Anthropic states the Max and Team multiples for the per-session allowance; it doesn't give a multiple for weekly limits (Max, Team). Claude Code isn't on the free plan (Is Claude Code free?).
How do Claude Code usage limits work?
- Session limit. "Your session-based usage limit will reset every five hours," Anthropic says (Pro).
- Weekly limit. Applies across all models and resets at a fixed time each week assigned to your account. Settings > Usage shows when.
- One pool. Anthropic says limits are "shared across Claude and Claude Code", so a long chat in the morning leaves less for coding in the afternoon (Anthropic).
- Bigger models cost more of it. Anthropic says Opus "uses meaningfully more of your quota" than Sonnet (Anthropic). Claude Code's error messages also include separate Opus and Sonnet limits, whose size isn't published.
- Fable models. On Max and premium seats, up to half the weekly limit can go to Fable 5 and 5.1. On Pro and Team standard seats, Fable runs on usage credits from the first request (Anthropic).
Anthropic also says it may limit usage in other ways at its discretion.
How do you check your Claude Code usage?
- In Claude Code, run
/usage. It shows your plan's limits, session cost and a breakdown of what used them, including skills, subagents and MCP servers (commands). - On claude.ai, open Settings > Usage. It shows the current session with the time left and your weekly limits.
- Near the limit, Claude Code warns you with the percentage used and the reset time, and on Pro and Max a custom status line can show both limits once the session's first reply arrives (Claude Code status line).
Since version 2.1.234, when a limit stops Claude mid-task, Claude Code waits in the open session and continues on its own after the reset (docs).
What can you do when you hit the limit?
- Wait for the reset. Five hours at most for the session limit.
- Turn on usage credits. On Pro and Max: Settings > Usage > Usage credits. You set a monthly spend limit and prepay, billed at standard API rates; in Claude Code the command is
/usage-credits, previously/extra-usage(Anthropic). On Team, owners buy credits and set spend limits for the organization, by seat type or by member. - Use a saved reset. Since September 22, subscribers get one limit reset to use when they choose, which refills either the session or the weekly limit. It's in Settings > Usage, not the terminal (Anthropic).
- Switch to an API key. A Claude Console account bills per token: Sonnet 5.5 at $2 per million input tokens and $10 output, Opus 5.5 at $4 and $20, Haiku 5.5 at $0.10 and $0.50 (Anthropic).
- Upgrade to Max, or to a premium seat on Team.
Fast mode runs only on usage credits. And the prompt cache lasts an hour on a subscription but five minutes on credits, so the same work can cost more once you're past the limit (costs).
What changed in 2026?
- May 6: Anthropic doubled Claude Code's five-hour limits on Pro, Max, Team and seat-based Enterprise, and removed the peak-hours reduction for Pro and Max (Anthropic).
- May 13: a temporary 50% increase in weekly limits, later extended until September 14 (Anthropic on X).
- September 14: weekly limits became permanently 25% above standard. Anthropic noted that compared with the temporary boost, this "works out to a 17% reduction" (Anthropic on X).
- September 22: with Opus 5.5, five-hour limits went up again on the same plans, by an unpublished amount, and subscribers got the saved limit reset (Anthropic).
- Enterprise moved to a $20 seat with all usage at API rates. Older seat-based Enterprise plans keep their limits until renewal (Anthropic).
How do you make the limits last?
Anthropic's own advice (help center, docs):
- Run
/clearbetween tasks. Anthropic calls it the single most effective lever. Use/compactpartway through a long task. - Use Sonnet for most coding and Opus when you're stuck.
/model opusplanplans with Opus and writes the code with Sonnet. - Point to file paths instead of pasting files.
- Keep CLAUDE.md under about 200 lines, and move the rest into skills.
- Lower the effort level with
/effortfor routine work. - Turn off MCP servers you aren't using.
The full list, with what each saves, is in how to reduce Claude Code token usage.
How do Claude Code limits compare with Codex?
OpenAI publishes ranges where Anthropic publishes multiples. On ChatGPT Plus and Business, Codex estimates local messages per five-hour window by model, from 350 to 3,000 for GPT-6 Luna down to 15 to 160 for GPT-6.1 Sol, and weekly limits may also apply. ChatGPT Pro currently has no five-hour limit (OpenAI). The full comparison is in Claude Code vs Codex.
Usage limits on a team
On Team, each member has their own limits, so a heavy user can't borrow a teammate's unused allowance; premium seats and owner-bought credits are the fixes (Claude Code for teams).
Poly (usepoly.co) runs Claude Code in a shared room instead: each turn is charged to the person who ran it, watching and approving are free, and nobody needs their own Claude plan or API key. Free to start. What is Poly?
Common questions
How many messages do you get with Claude Code on Pro?
Anthropic doesn't publish a number. Pro has a session limit that resets every five hours and a weekly limit, and how far they go depends on the model, the size of your codebase and how long your conversations run.
When do Claude Code usage limits reset?
The session limit resets every five hours. The weekly limit resets at a fixed time each week assigned to your account, shown in Settings > Usage.
Do Claude chat and Claude Code share usage limits?
Yes. Anthropic says limits are shared across Claude and Claude Code, so all activity in both counts against the same allowance.
What is the difference between Claude Max and Pro usage limits?
Max 5x includes five times Pro's per-session allowance and Max 20x twenty times. Both also have weekly limits, and Max can spend up to half of its weekly limit on Fable models.
How do you increase your Claude Code usage limit?
Turn on usage credits, billed at API rates, use a saved limit reset, switch to an API key, or upgrade to Max or a Team premium seat.
Are Claude Team usage limits shared across the team?
No. Team limits are per member. Owners can buy usage credits for the organization and set spend limits by seat type or by member.