No. The software is free to install, but it needs a paid Claude plan or API credits to run. The free Claude.ai plan doesn't include Claude Code. The cheapest option is Claude Pro at $17 a month billed annually or $20 billed monthly.

How much does Claude Code cost?

As of October 2026, Claude Code is included in Pro ($20 a month, or $17 billed annually), Max 5x ($100 a month), Max 20x ($200 a month), Team ($20–25 per standard seat, $100–125 per premium seat) and Enterprise ($20 per seat plus usage at API rates). You can also pay per token through the Claude API. Anthropic reports an average of about $13 per developer per active day for enterprise API usage.

Is Claude Max worth it for Claude Code?

For heavy daily use, usually yes. Max 5x gives five times Pro's usage per five-hour session for $100, and Max 20x gives twenty times for $200. Anthropic's enterprise data puts typical API spend at $150–250 per developer per month, so a full-time user who would otherwise pay API rates often comes out ahead on Max.

What happens when I hit my Claude Code usage limit?

Limits reset on a rolling five-hour window, and paid plans also have weekly limits. When you hit one, you can wait for the reset, upgrade your plan, or turn on usage credits to keep working at standard API rates.

Do Claude chat and Claude Code share the same limits?

Yes. On subscription plans, usage in Claude on web, desktop and mobile and in Claude Code all draws from the same pool.

Can I use Claude Fable in Claude Code?

Yes, but not on every plan. As of October 17, 2026, Max plans include Fable at up to 50% of the weekly usage limit, Team premium seats include it at up to 50% of each member's weekly limit, and Enterprise includes it. On Pro, Fable bills to usage credits at API rates. Fable is never the default model, so pick it with /model fable. On the API it costs $10 per million input tokens and $50 per million output tokens, 2.5 times Opus 5.5.

Can I use Claude Code with an API key instead of a subscription?

Yes. Set ANTHROPIC_API_KEY from a Claude Console account, or point Claude Code at a compatible API gateway. You then pay per token instead of a monthly fee, with no five-hour limits, but costs scale with how much you use it.

Claude Code Pricing (2026): Is It Free? Pro vs Max vs API Costs Explained

Short answer: Claude Code is not free. The free Claude plan doesn't include it. It comes with every paid plan: Pro $20/month ($17 billed annually), Max $100 or $200/month, Team from $20 per seat, or Enterprise at $20 per seat plus API usage. You can also skip the subscription and pay per token through the Claude API, where Anthropic says enterprise developers average about $13 per active day.

Prices below are from Anthropic's pricing page and Claude Code documentation, checked on October 17, 2026, in USD and excluding tax.

Plans that include Claude Code

Plan	Price	Usage (vs. Pro)	Extras	Good for
Free	$0	—	No Claude Code	Chat only
Pro	$20/mo, or $17/mo billed annually ($200/yr)	Baseline		Trying it, a few sessions a week
Max 5x	$100/mo	5x per 5-hour session	$100/mo in API credits, priority access	Daily use
Max 20x	$200/mo	20x per 5-hour session	$200/mo in API credits, priority access	All-day, heavy use
Team, standard seat	$20/seat/mo annual, $25 monthly	More than Pro	Admin, SSO, up to $500/mo pooled API credits	Teams of 2–150
Team, premium seat	$100/seat/mo annual, $125 monthly	5x a standard seat		Heavy users on a team
Enterprise	$20/seat/mo (annual, 20-seat minimum) + usage at API rates	Usage-based	SCIM, audit logs, spend controls	Large organizations

Max only bills monthly. Annual discounts apply to Pro and Team.

How the usage limits work

Subscription plans don't give you a fixed number of messages. Instead:

Limits reset on a rolling 5-hour window, and paid plans also have weekly limits.
Everything shares one pool. Chats in the Claude app and work in Claude Code count against the same limits.
How far a plan goes depends on the model and the size of your context. Opus burns through a window faster than Sonnet, and a long session costs more than a short one.
When you hit a limit, you can wait, upgrade, or turn on usage credits to keep going at standard API rates. Run /usage-credits inside Claude Code to manage them.

Inside Claude Code, /usage shows your plan's usage bars, and it flags anything eating a lot of your allowance, such as long context or cache misses.

Claude Fable in Claude Code

Claude Fable 5.1 and Fable 5 are Anthropic's most capable models, built for long, autonomous tasks. Whether your plan covers them depends on the tier:

Plan	Fable
Free	No (Free doesn't include Claude Code at all)
Pro	Only through usage credits, billed at API rates
Max 5x and Max 20x	Included, up to 50% of your weekly limit
Team	Premium seats: included, up to 50% of each member's weekly limit
Enterprise	Included

It's opt-in. Neither Fable model is the default on any plan. Run /model fable for Fable 5.1, or start with claude --model fable. For Fable 5, use its model ID: /model claude-fable-5.
Watch the usage credits. When Fable would bill to usage credits, the /model picker shows "Requires usage credits", and an interactive session asks before the first such request. Non-interactive runs (claude -p) and Agent SDK apps don't ask: Claude Code bills the credits straight away.
It's the most expensive model by far. On API rates Fable costs $10 / $50 per million input / output tokens, 2.5 times Opus 5.5. Its thinking can't be turned off and bills as output. Anthropic suggests starting with Opus 5.5 and moving up to Fable when a task needs it. Per-token details are in Claude API pricing.

Paying per token instead (Claude API)

With a Claude Console API key, Claude Code bills every token at API prices. There's no 5-hour window, just a bill that grows with use. Current prices for the models Claude Code uses most:

Model	Input	Output	Cache read
Claude Sonnet 5.5	$2 / MTok	$10 / MTok	$0.10 / MTok
Claude Opus 5.5	$4 / MTok	$20 / MTok	$0.20 / MTok
Claude Haiku 5.5	$0.10 / MTok	$0.50 / MTok	$0.01 / MTok
Claude Fable 5.1	$10 / MTok	$50 / MTok	$0.25 / MTok

What that adds up to:

Anthropic's numbers: across enterprise deployments, about $13 per developer per active day and $150–250 per developer per month, with 90% of users under $30 per active day.
Our estimate for one long session: a 50-turn Opus 5.5 session averaging 60K tokens of context per turn comes to about $2.60 with prompt caching (Claude Code caches automatically), and about $13.50 if caching weren't working. The full calculation is in Claude API pricing.

Extended thinking is always on for Opus 5.5, Sonnet 5.5, Haiku 5.5 and the Fable models, and thinking tokens bill as output. Lower the effort level with /effort for simple tasks.

Subscription or API: how to decide

You…	Best fit	Why
Code with Claude a few times a week	Pro	$20 covers occasional sessions
Use it every working day	Max 5x	$100 is below what an average enterprise developer spends at API rates ($150–250/month)
Run long agent sessions all day	Max 20x	$200 a month, flat
Run it in shared CI or on servers	API key	Per-use billing and a key you can revoke. A subscription token from `claude setup-token` also works, but it draws on your personal limits
Need central billing, SSO and spend controls	Team or Enterprise
Can't pay Anthropic directly	Cloud provider or API gateway	See below

A useful test: run /usage for a week on Pro. If you keep hitting the 5-hour limit, Max is the cheaper upgrade. If you'd rather never think about limits, an API key is the most flexible option, and the costs are easy to track.

Other ways to pay for Claude Code

Amazon Bedrock, Google Cloud, Microsoft Foundry: Claude Code can use Claude models from these clouds, billed to your cloud account.
API gateways: Claude Code supports routing through a gateway with ANTHROPIC_BASE_URL and the gateway's key. api.foo, which publishes these docs, is one option. See Install Claude Code → Use an API gateway. Note that subscription usage limits don't apply to gateway traffic. You pay the gateway's per-token rates instead.

How to keep Claude Code costs down

Use Sonnet 5.5 for routine work. Check which model you're on with /model. Save Opus for hard problems and Fable for long autonomous tasks: per token, Opus costs twice as much as Sonnet and Fable five times as much, and both use up plan limits faster.
/clear between unrelated tasks. Stale context is re-sent, and billed, on every turn.
/compact long sessions and tell it what to keep: /compact Focus on the failing tests.
Keep CLAUDE.md short and move occasional workflows into skills, which load only when used.
Turn off MCP servers you aren't using (/mcp), and check what fills your context with /context.
Lower the thinking effort (/effort) for routine edits.