Claude Pro costs $20 per month or $200 per year, equivalent to $16.67 per month when paid annually. Claude Code is still included in Pro in 2026. Developers who need more capacity can choose Max 5x at $100 per month or Max 20x at $200 per month, while API users pay separately for the tokens they consume.

Pro includes Claude Code, but Claude chat and coding sessions share usage limits. This breakdown compares Pro, Max and API billing, including what happens when repository work exhausts the included allowance.

Pricing confirmed on 18 June 2026. Prices are in US dollars before applicable taxes and regional adjustments.

Claude Code pricing at a glance

Route	Price	What you get	Best fit
Claude Pro	$20 monthly or $200 yearly	Claude Code with shared subscription limits	Occasional coding and smaller repositories
Max 5x	$100 monthly	Five times Pro usage per session	Frequent daily development
Max 20x	$200 monthly	Twenty times Pro usage per session	Heavy agent workflows
Claude API	Pay per token	Metered model access through Console billing	Automation and controlled workloads

Max mainly buys more capacity, higher output limits and priority access. API billing remains separate, so Pro or Max does not provide a general API allowance for your own applications.

---

---

Is Claude Code still included in Claude Pro?

Yes. Claims that Claude Code moved out of Pro and now requires Max are incorrect as of June 2026. Anthropic still lists Claude Code as a Pro feature and allows Pro subscribers to authenticate in the terminal or via a supported IDE extension using their Claude account.

The confusion comes from the Pro’s lower allowance. Focused fixes and short repository sessions may fit comfortably, while multi-file work with repeated test runs can reach the shared limit quickly.

Claude and Claude Code draw from the same capacity, so heavy use of either product reduces what remains for the other.

How Pro and Max usage limits work

Pro is not sold as a fixed number of tokens. Usage varies with the model, repository size, conversation history, tool calls and output length. Limits include rolling five-hour controls.

After reaching the allowance, you can wait, upgrade, or enable prepaid usage credits, which are charged at standard API rates. Keep extra usage disabled or capped when a fixed monthly bill matters.

Anthropic’s Claude Code billing guidance also warns that an ANTHROPIC_API_KEY environment variable can cause Claude Code to use API billing instead of the included subscription. Check the active login and use /status it before a long session.

Claude API pricing and real coding costs

Model	Input per million tokens	Output per million tokens
Claude Haiku 4.5	$1	$5
Claude Sonnet 4.6	$3	$15
Claude Opus 4.8	$5	$25

API input includes prompts, code, repository context and previous conversation content. Output includes plans, explanations and generated patches. One million Sonnet 4.6 input tokens plus 50,000 output tokens costs about $3.75. The same volume on Opus 4.8 costs about $6.25.

Agent loops can resend large contexts over several steps. API access suits scripts, CI jobs and measurable bursts; subscriptions are easier to budget for regular interactive use.

Claude Code vs Cursor vs GitHub Copilot on price

Tool	Entry paid plan	DIY AI rating	Main trade-off
Claude Code	$20 monthly	9.2/10 – 4.6/5	Best repo reasoning, but shared limits can interrupt heavy sessions
Cursor	$20 monthly	9.1/10 – 4.6/5	Stronger IDE workflow, with larger plans needed for heavy agents
GitHub Copilot	$10 monthly	9.0/10 – 4.5/5	Lower entry price, but agent work consumes AI credits

Claude Code ranks first in the DIY AI code-generation dataset for repository context, refactoring and test generation. Cursor favours an editor-led workflow, while GitHub Copilot costs less for lighter assistance.

See the Claude Code free-tier explanation for access and installation questions. Our Claude Code best-practices guide covers reducing wasted context and verifying edits. For complete site creation, compare AI coding tools for building websites.

Claude Code pricing pros and cons

Pros	Cons
Claude Code is included in the $20 Pro subscription.	Usage is the same as in normal Claude conversations.
Annual Pro billing lowers the effective monthly cost.	The jump from Pro to Max 5x is substantial.
Max can beat API billing for sustained heavy use.	Plans do not provide a simple fixed token allowance.
API billing supports automation and model routing.	Long agent loops can create unpredictable costs.

Verdict: Pro, Max or API?

Start with Pro when Claude Code is an occasional assistant rather than the centre of the working day. Upgrade to Max 5x only when Pro limits repeatedly interrupt useful paid work. Max 20x is for developers who already know that the 5x allowance is insufficient.

Choose API billing for measurable automation and integrations. Most individual developers should start with Claude Pro, monitor a typical month, and upgrade only when limits repeatedly disrupt their work.

Claude Code pricing FAQs

How much does Claude Pro cost?

Claude Pro costs $20 per month or $200 per year. The annual option works out at approximately $16.67 per month, and Claude Code remains included.

Is Claude Pro worth the cost?

Pro is a good value for regular Claude users who need shorter coding sessions and can work within shared limits. Heavy daily Claude Code users are more likely to benefit from Max.

Is Claude cheaper than ChatGPT?

Claude Pro and ChatGPT Plus both cost $20 per month. Claude offers discounted annual Pro billing, but the coding value depends on whether Claude Code or Codex better fits the workflow and usage level.