Claude Code Pricing (October 2026): Which Plans Include It, Usage Limits and Pay-per-Token Billing

Quick answer: Claude Code has no separate price. It is included in Claude Pro, Max 5x, Max 20x, Team and Enterprise, and it is not on the Free plan. The other route is pay per token through a Claude Console (API) account. Start with Pro, move to Max if you hit limits every day, and use the API route for automation or when you want a hard spend cap. Current prices are on claude.com/pricing (accessed October 25, 2026).

Which plans include Claude Code?

Plan	Claude Code?	Usage and billing differences	Official pricing
Free	No	Chat only	claude.com/pricing
Pro	Yes	Entry paid plan; monthly or cheaper annual billing	claude.com/pricing
Max 5x	Yes	5x Pro usage; monthly billing; priority access at busy times	claude.com/pricing
Max 20x	Yes	20x Pro usage; monthly billing; larger monthly API credit allowance	claude.com/pricing
Team (Standard seat)	Yes	Teams of 2-150; more usage than Pro	claude.com/pricing
Team (Premium seat)	Yes	5x the usage of a Standard seat; mix seat types	claude.com/pricing
Enterprise	Yes	Seat price plus usage at API rates; admin spend limits	claude.com/pricing
Claude Console (API)	Yes	Pay per token, no subscription; workspace spend limits	API pricing

Claude Code also runs on Amazon Bedrock, Google Cloud’s Agent Platform and Microsoft Foundry, billed per token to your cloud account (Claude Code cost guide).

Subscription or pay per token: which costs less?

It depends on how much you use it:

• Light or bursty use: pay per token can be cheaper, because you pay nothing on days you don’t code. Run /usage to see the token cost of a session.
• Daily, all-day use: a subscription is usually better value. Pro and Max subscribers have usage included, so Anthropic notes the session cost figure in /usage “isn’t relevant for billing purposes”.
• A team: on Team and Enterprise, each member’s Claude Code usage draws from a per-seat allowance, and admins can let members continue past it with usage credits and spend limits.

Anthropic’s cost guide also gives average per-developer spend figures for enterprise deployments. Use them only as a starting point: it recommends piloting with a small group to set your own baseline.

For a full plan-by-plan breakdown, including what Max’s API credits cover, see Claude Pro vs Max vs API.

How do Claude Code usage limits work?

On subscription plans, Claude Code shares one allowance with Claude chat and Cowork. The allowance resets on a rolling five-hour window and a weekly window (cost guide). Messages to know:

• “You’ve hit your session limit” or “weekly limit”: applies across all models, so switching with /model won’t help. Wait for the reset time in the message, or turn on usage credits.
• “You’ve hit your Opus limit” or “Sonnet limit”: model-specific; switching to another model family keeps you working.
• Spend-limit messages: you are paying with usage credits and reached a spend limit you or your admin set.

To keep working past your plan’s limit, run /usage-credits (signed in with your claude.ai subscription). On Pro and Max it opens Settings > Usage, where you turn usage credits on and set a monthly spend limit.

Why does Claude Code use up limits so fast?

Claude Code sends your full conversation with every request, so cost grows with context. The main drains, per Anthropic’s cost guide:

• Long sessions that are never cleared.
• Cache misses after a break longer than the cache lifetime.
• Subagents, agent teams and scheduled tasks, which send their own requests. Anthropic says agent teams use about 7x more tokens than standard sessions when teammates run in plan mode.
• Extended thinking, billed as output tokens.

How to cut usage:

1. Run /usage for a breakdown of what is using your plan, with suggested fixes.
2. Use /clear between unrelated tasks (it costs nothing; /compact itself is a large request).
3. Pick the model for the job: Sonnet for most coding, Opus for hard design decisions, Haiku for simple subagents.
4. Lower the effort level with /effort for simple tasks.
5. Keep CLAUDE.md under about 200 lines and move specialized instructions into skills.
6. Disable unused MCP servers with /mcp, and prefer CLI tools like gh where they exist.

How do you control Claude Code spend on the API?

When you first sign in to Claude Code with a Claude Console account, the Console creates a workspace called “Claude Code” for tracking all Claude Code usage. You can’t create API keys in it, but admins can set workspace spend limits and a workspace rate limit there. To get a Console account and key, see how to get a Claude API key, and check per-token rates in Claude API pricing.

More guides are in the Claude hub.