How much does Claude Code cost?
Claude Code has no separate price: it comes with every paid Claude plan, and as of October 2026 the cheapest is Pro at $20 a month ($17 a month billed annually), per claude.com/pricing. The free Claude plan does not include it. If you would rather not subscribe, you can connect a Claude Console account and pay per token at API rates.
Anthropic's pricing FAQ puts it plainly: "Claude Code is included in all paid plans. It shares the same usage limits as the rest of your plan, so your work in the terminal and your chats draw from one pool." So the real question is not the price of Claude Code, but which plan gives you enough usage for the way you work.
All prices on this page were read from Anthropic's official pages on 23 October 2026. Prices exclude tax and change from time to time, so check the linked page before you buy.
Which Claude plans include Claude Code?
| Plan | Price (as of October 2026) | Claude Code | Usage, as Anthropic describes it |
|---|---|---|---|
| Free | $0 | Not included | Covers "everyday questions" |
| Pro | $20/month monthly, or $17/month billed annually ($200 up front) | Included | At least 5x Free per 5-hour session |
| Max 5x | $100/month | Included | 5x Pro per 5-hour session |
| Max 20x | $200/month | Included | 20x Pro per 5-hour session |
| Team, Standard seat | $20/seat/month annually, $25 monthly | Included | More than Pro |
| Team, Premium seat | $100/seat/month annually, $125 monthly | Included | 5x Standard seats |
| Enterprise | $20/seat/month billed annually, 20-seat minimum, plus usage at API rates | Included | Scales with model and task |
Sources: claude.com/pricing for Pro, Team and Enterprise, and Anthropic's Max plan article for the two Max prices, which are monthly only. Team plans are for 2 to 150 people.
Pro vs Max: which do you need?
Pro is the entry point. It includes Claude Code, Opus and Sonnet, and the rest of the Claude apps. If you code with Claude Code for focused stretches a few times a week, start here.
Max is for people who use Claude all day. Anthropic's pricing page describes it as "for people who work with Claude all day," with 5x or 20x Pro's usage per five-hour session, higher output limits and priority access at busy times. Max also comes with $100 or $200 a month in API credits for building on the Claude Platform, but Anthropic's Claude Code support article says those monthly API credits "can't be used in Claude Code or the Claude apps, including for extra usage."
A note on Fable, Anthropic's most capable model in Claude Code. It is never the default; you opt in with /model fable. Depending on your plan and seat tier, Fable can bill to usage credits rather than your included limits, and Max has a separate weekly limit for Fable. The model configuration docs explain that Claude Code shows "Requires usage credits" in the /model picker when that applies to you.
A simple rule: start on Pro. If you hit the five-hour limit most days, move to Max 5x. If you still hit it, move to Max 20x. Upgrades are prorated.
Team and Enterprise seats
On Team, every seat includes Claude Code. Standard seats give more usage than Pro, and Premium seats give five times a Standard seat, so a team can mix seat types and give Premium only to the heaviest users. Team seats sign in to Claude Code by choosing "Claude account with subscription" at login, per Anthropic's Team and Enterprise article.
Enterprise works differently: a per-seat price plus usage billed at API rates, so cost follows what people actually run. On usage-based Enterprise, Anthropic says "there are no per-seat usage limits." Admins can turn on usage credits and set spend limits at the organization, group or member level.
For budgeting, Anthropic's cost docs say that across enterprise deployments "the average cost is around $13 per developer per active day and $150-250 per developer per month, with costs remaining below $30 per active day for 90% of users." Treat that as a starting point, not a quote; your own pilot will tell you more.
Paying per token with the API instead
You can skip the subscription and use Claude Code with a Claude Console account. Every token is billed at API rates, there are no plan limits, and spend shows up on the Console usage page. When you first sign in this way, Claude Code creates a workspace called "Claude Code" for cost tracking.
API prices per million tokens, as of October 2026, from claude.com/pricing:
| Model | Input | Output |
|---|---|---|
| Fable 5.1 | $10 | $50 |
| Opus 5.5 | $4 | $20 |
| Sonnet 5.5 | $2 | $10 |
| Haiku 5.5 (prompts up to 100K tokens) | $0.10 | $0.50 |
Cached input costs much less than fresh input (for Opus 5.5, cache reads are $0.20 per million tokens), and Claude Code uses prompt caching automatically.
One trap to know about: if the ANTHROPIC_API_KEY environment variable is set, Claude Code uses that key instead of your subscription, and you are billed for API usage. If you meant to use your plan, unset it and run /login.
Claude Code usage limits
Anthropic does not publish a fixed message count for any plan. The pricing FAQ explains that usage depends on "your conversations, the model you choose, and the features you use." What it does describe:
- Five-hour session limit. Pro and Max both have one. Max gives 5x or 20x Pro's allowance per session.
- Weekly limit. Pro and Max also have a weekly limit. On Max it resets at a fixed time assigned to your account, and there is a separate weekly limit for Fable.
- Shared pool. Claude chat, Claude Code in the terminal and Claude Code in VS Code, Cursor or JetBrains all count against the same limits.
- Other caps. Anthropic says it may also limit usage "in other ways, such as weekly and monthly caps or model and feature usage."
When you hit a limit, you can wait for the reset, move up a plan, or turn on usage credits, which let you keep working "at standard API rates." Claude Code may also let a response that is already running finish its current step, with the message "Usage limit reached · wrapping up."
How to see your usage and costs
Claude Code has one main command for this:
/usage
On Pro, Max, Team or Enterprise, /usage shows your plan usage bars and a breakdown of what counted against your limits: skills, subagents, plugins and individual MCP servers, with suggestions such as trimming a large CLAUDE.md. Press d or w to switch between the last day and the last week, and press f to have Claude Code apply the recommendations it can. With an API account, the Session block shows an estimated dollar cost for the session. /cost and /stats open the same screen, per the commands reference.
Other places to look:
/contextshows what is filling the context window right now./usage-creditsopens your usage credit settings on claude.ai.- Settings > Usage on claude.ai shows where you stand across all Claude apps.
- The status line can show context use or session cost all the time; see the status line guide.
- The Claude Console usage page is the authoritative bill for API accounts.
Ways to spend less with Claude Code
Most of the cost of a turn is the context sent with it, so the biggest savings come from keeping context small. These come from Anthropic's cost docs:
- Pick the right model. The docs say Sonnet "handles most coding tasks well and costs less than Opus." Switch with
/model sonnet, and give simple subagentsmodel: haiku. - Lower the effort. For simple work,
/effort lowor/effort mediumcuts thinking tokens, which are billed as output. - Clear between tasks.
/clearstarts fresh at no cost. Stale context is paid for again on every message. - Compact with a focus.
/compact Focus on code samples and API usagekeeps what matters. Compacting a big conversation is itself a large request, so don't run it constantly. - Keep CLAUDE.md lean. The docs suggest keeping it under 200 lines and moving step-by-step workflows into skills, which load only when needed. See the CLAUDE.md guide.
- Turn off unused MCP servers. Run
/mcpand disable what you aren't using. CLI tools likeghare lighter still. - Plan before big changes. Plan mode catches a wrong direction before Claude writes code you will throw away.
- Be specific. "Add input validation to the login function in auth.ts" costs far less than "improve this codebase."
How Codex and Cursor price, briefly
- Codex comes with ChatGPT plans. As of October 2026, OpenAI's Codex pricing page lists the CLI and IDE extension on Plus ($20 a month), Pro (from $100 a month) and Business ($20 per user per month billed annually), with usage limits that depend on the model and the task. You can also sign in with an OpenAI API key and pay standard API rates.
- Cursor sells plans with a pool of AI usage that, in its words, "refreshes weekly." As of October 2026, cursor.com/pricing lists Premium at $30 a month, Super at $100 and Ultra at $200. Premium and above don't include an allowance for third-party models; those are billed separately at API list price, and you can set spending caps.
So all three follow a similar pattern: a monthly plan with usage limits, plus a pay-per-token path for heavy or automated use. For a feature comparison, see Codex vs Claude Code and Claude Code vs Cursor.
Get more out of every plan
Whatever you pay, reusable instructions stretch your limits further: a good skill saves Claude from rediscovering your project every session. Kipt keeps your prompts and skills as Markdown in a private GitHub repo you own and installs them into Claude Code with one command. The free SKILL.md generator turns a prompt you keep retyping into a skill.