Claude Code comes with a paid Claude plan: Pro at $20 a month, Max at $100 or Ultra at $200. The free plan does not include it. You can also pay per token with an API key, which Anthropic says averages about $13 per developer per active day in company use.
Plans for one person
| Plan | Price | Claude Code | Usage |
|---|---|---|---|
| Free | $0 | Not included | |
| Pro | $20 a month, or $17 a month paid yearly ($200) | Included | A 5-hour limit and a weekly limit |
| Max | $100 a month | Included | 5 times Pro, every 5 hours and every week |
| Ultra | $200 a month | Included | No 5-hour limit (verified accounts); 10 times Pro each week |
The plan price is all you pay unless you choose to go past your limits. Usage inside the limits is not billed per token. How the 5-hour and weekly limits work.
Plans for a team
| Plan | Price per seat |
|---|---|
| Team, standard seat | $25 a month, or $20 paid yearly |
| Team, premium seat | $125 a month, or $100 paid yearly |
| Enterprise | $20 a month paid yearly (20 seat minimum), plus usage at API rates |
Team is for 2 to 150 people and lists Claude Code among its features.
Paying per token with an API key
With a Claude Console account, Claude Code bills for the tokens it uses. Prices are per million tokens:
| Model | Input | Cached input (read) | Output |
|---|---|---|---|
| Claude Fable 5.2 | $8 | $0.20 | $40 |
| Claude Opus 5.5 | $4 | $0.20 | $20 |
| Claude Sonnet 5.5 | $2 | $0.10 | $10 |
| Claude Haiku 5.5 (prompts up to 100,000 tokens) | $0.10 | $0.01 | $0.50 |
| Claude Haiku 5.5 (longer prompts) | $0.50 | $0.05 | $2.50 |
Writing to the cache costs a little more than plain input (1.25 times for a 5-minute cache). Most of what Claude Code sends on each turn is the conversation so far, read from the cache, so the cached price is the one that matters most in a long session.
For a sense of scale, Anthropic's documentation says that across company deployments the average is about $13 per developer per active day, $150 to $250 per developer per month, and under $30 per active day for 90% of users.
Going past a plan's limits
When you hit a limit you can wait for it to reset, upgrade, or turn on usage credits, which let you keep working with the extra billed at standard API rates up to a monthly cap you choose.
Plan or API key?
- A plan suits steady daily use. The cost is fixed and known.
- An API key suits light or occasional use, or automation, where you would not use a plan's allowance.
- If you use it heavily, a plan is usually far cheaper than the same tokens at API prices. The catch is the limits: the question stops being "what did it cost" and becomes "will I run out".
What decides how much you use
Three things, on a plan or a key: how long your sessions get, how much loads before any work, and how much is done by subagents. Why Claude Code gets expensive, and the fixes.
Cost Doctor shows which of your sessions is using the most and why, with a forecast of when your plan limit runs out. The free costdoctor command prints the same in a terminal.
Prices on this page are from Anthropic's plans page, API pricing page and Claude Code cost documentation, as checked on November 10, 2026, in US dollars before tax. Anthropic says prices and plans are subject to change; check those pages for the current ones.
See it on your own sessions
Cost Doctor is a Mac menu bar app that reads the logs Claude Code and Codex already write. It names the session that is using your tokens, says why, and turns your screen red when a check fails while an agent sits waiting. Nothing leaves your Mac.
Download the free trialHow it works14 days free, no account. Then $19 once, plus tax where it applies. For Apple silicon and Intel Macs on macOS 13 or later.