SwitchEarly accessClaude Code usage limits explained
Claude Code usage limits are your Claude plan's limits. On Pro, Max, Team and seat-based Enterprise plans, Claude Code shares the same five-hour session limit and weekly limit as chat, so all your activity draws from one pool. If you sign in with an API key instead, there's no plan limit and you pay per token.
A guide from Switch, an independent Claude Desktop profile manager. Switch is in early access ahead of its public release.
How limits apply to Claude Code
As of October 2026, Anthropic says Pro and Max plans have a five-hour session limit and a weekly limit, and Max plans also have a separate weekly limit for Fable. These limits are shared across Claude and Claude Code. That includes Claude Code in Claude Desktop, in the terminal and in supported IDEs (Use Claude Code with your Pro or Max plan).
How you signed in decides which rules apply (Models, usage, and limits in Claude Code):
- A claude.ai plan (Pro, Max, Team, Enterprise seat): included usage, reset on a rolling window. Running out shows a "limit reached, resets at" message.
- An API key (Console or a cloud provider): pay-as-you-go per token, with no hard stop.
What happens when you hit the limit
You'll see warnings as you get close. If you reach the session limit while Claude Code is partway through a response, it may finish the step in progress and show "Usage limit reached · wrapping up". Anthropic says this allowance is small and counts toward your weekly limit. On Pro it applies up to once per week; on Max it can apply at each session limit. It is rolling out gradually.
After that, Anthropic lists these options: wait for the reset, turn on usage credits, upgrade (Pro to Max 5x, or Max 5x to Max 20x), or switch to a Console account and pay API rates for heavy periods. Max's monthly API credits can't be used in Claude Code.
Make your allowance last longer
Anthropic's Claude Code guidance says most "used it all by lunchtime" cases come from a few habits:
- Clear between tasks. Every earlier message is sent again with each turn, so run
/clearwhen you change task, or/compactto keep a summary. - Match the model to the job. Opus uses noticeably more of your quota than Sonnet. Use
/modelto switch. - Point at files instead of pasting them. Pasted text stays in context for the rest of the session.
- Keep CLAUDE.md short. It's included with every turn.
- Ask for a plan before big changes, so you don't pay for a large diff you then throw away.
Run /usage to see which skills, subagents, plugins and MCP servers used the most (Claude Code costs).
Keeping work and personal usage separate
Many developers have two accounts they're entitled to: a Team or Enterprise seat from an employer, and a personal Pro or Max plan. Keeping them separate means work usage counts against the work seat and personal projects against your own plan, and each account's limits stay its own.
Claude Code supports this. Anthropic's authentication docs say each CLAUDE_CONFIG_DIR has its own settings, session history and login, so you can stay signed in to both. Run multiple Claude Code accounts on one Mac shows the setup. If you use Claude Code inside Claude Desktop, Switch gives each profile its own Claude Code configuration folder when "Keep coding preferences separate" is on (the default for new profiles); see Switch for Claude Code.
Separate accounts don't add up limits
To be plain about it: a second account is not a way to get more Claude Code usage. Each account keeps its own plan and limits, Team limits are per named member and can't be shared, and rotating accounts to get around limits can conflict with Anthropic's terms. If you keep running out on one account, the supported options are a larger plan, usage credits or API billing. Subscription vs API billing compares the last two.