Fix Claude Code usage/rate limit reached

Q: How do I see how much Claude Code usage I have left?

Run /usage. On a Pro, Max, Ultra, Team, or Enterprise plan it shows your plan usage bars, a breakdown of what counts against your limits, and when they reset.

Q: Does /clear or /compact save usage?

Use /clear when you switch to unrelated work, because stale context wastes tokens on every later message. /compact summarizes the conversation, but it reads the whole history to do so, so compacting a large context is itself a large request.

A usage or rate limit message in Claude Code means your account spent its allowance, not that anything broke. On a subscription you wait for the named reset, switch model family if the limit is model-specific, or turn on usage credits; on an API key it is a per-token rate limit. Here is what each means and how to resume.

1. Read which limit you hit

You've hit your session limit

A session or weekly limit is seat-based and shared across every model, so switching models will not help; the message names its reset time.

You've hit your Opus limit

A model-specific limit is different: another family keeps working.

2. Wait or auto-resume after reset

/rate-limit-options

Open this to have Claude Code resume the interrupted task automatically after the reset, or wait until the time it named.

3. Switch model for a per-model limit

/model sonnet

For an Opus or Sonnet limit, switch family and keep going.

4. Shrink context to spend less

/compact

Compact the conversation to cut the tokens each turn sends, or run /clear to start fresh; see manage context.

5. Subscription vs API-key billing

/usage

On a subscription, usage draws from a per-seat allowance that resets on a rolling session window and a weekly window; /usage-credits requests more when credits are on. On an API key, every token is billed and you hit per-token rate limits instead. Anthropic does not publish fixed numbers and they change, so check current limits on the pricing page and in usage limits.

Verify it worked

Run /usage and confirm your plan bars are below the limit, or send a prompt and confirm it runs.

If it fails:

The limit returns right after the reset — a long session reprocesses its whole context; /clear between tasks.
Unknown command: /usage-credits — your Claude Code is too old; run claude update.
Switching models changes nothing — you hit a session or weekly limit, not a model limit; wait for the reset. The official cost and limit docs cover each case.

Questions people ask

Does switching models help when Claude Code hits a usage limit?

Only for a model-specific limit. The session and weekly limits are shared across all models, so switching does not restore access. The Opus and Sonnet limits each apply only to that model family, so /model sonnet keeps you working after an Opus limit.

How do I see how much Claude Code usage I have left?

Run /usage. On a Pro, Max, Ultra, Team, or Enterprise plan it shows your plan usage bars, a breakdown of what counts against your limits, and when they reset.

Are Claude Code limits shared with the Claude app?

Yes. Pro and Max plans have a five-hour session limit and a weekly limit, shared across Claude and Claude Code, so chat activity counts against the same allowance. Team and Enterprise seats draw from a per-seat allowance on a rolling five-hour window and a weekly window, shared with Claude chat and Cowork.

Does /clear or /compact save usage?

Use /clear when you switch to unrelated work, because stale context wastes tokens on every later message. /compact summarizes the conversation, but it reads the whole history to do so, so compacting a large context is itself a large request.