Claude Code Pricing in 2026: Every Plan, Real Limits, and What It Actually Costs

For a solo builder, start with Pro for Claude Code, then consider Max 5x when included usage repeatedly interrupts paid work. Choose Team when you need shared billing and administration. Monthly API credits on Max and Team fund eligible apps and agents, not interactive Claude Code or extra usage; the five-hour and weekly limits still matter.

For interactive coding, most solo builders want Pro first, and the moment Claude Code starts stopping you mid-task most days, Max 5x is the next tier to assess. If you also build apps and agents on the API, the included credits change that part of your budget. The plans below were re-verified against Anthropic's live pricing page, Claude Code's cost documentation, and Anthropic's support, model, and release pages on October 28, 2026. Listed web prices are in US dollars before applicable tax; mobile prices can differ.

The Team credit amounts are contributions to one organization balance, not separate spendable balances for each member. Anthropic's monthly API credit rules explain the seat calculation and claim requirements.

A quick note on what "Claude Code" even is here, because it changes how you read the price. Claude Code is Anthropic's coding agent that runs in your terminal or IDE, the thing you hand a task and it edits files, runs commands, and iterates. There is no separate flat-rate Claude Code subscription. The current paid Claude lineup includes it alongside the chat app, while API use is metered separately. That bundling drives most of the questions below.

Is Claude Code free, and is it still on Pro?

No. Claude's Free plan does not include Claude Code subscription access. Anthropic's current pricing FAQ says Claude Code is included in all current paid plans, and Pro is the lowest-priced individual subscription that includes it.

The official support article, Use Claude Code with your Pro or Max plan, confirms that one subscription gives you Claude on web, desktop, and mobile and Claude Code in your terminal. Pro includes it.

Every plan, what it costs, what you get

Pro is the first plan to try for a founder validating an idea or an indie hacker shipping a small product. Its annual option is paid upfront, not month by month at the advertised equivalent. Exact capacity varies with conversation length, model, effort, and features, so Anthropic does not promise a fixed number of coding hours.

Max 5x buys more per-session capacity. This is the tier to assess when Pro interrupts you too often, particularly on larger repositories where a single task consumes more context. Its multiplier describes the session allowance, not a published weekly token budget.

Max 20x is for workloads that still exhaust Max 5x. Anthropic's Max plan support page confirms both tiers and monthly-only billing. Neither removes weekly limits, so do not treat a higher tier as a guarantee of uninterrupted all-day agent runs.

Claude Code defaults to Opus 5.5 at medium effort on Pro, Max 5x, Max 20x, Team Standard, Team Premium, Enterprise, and direct Anthropic API accounts. That account-type default applies from Claude Code 2.1.280; your saved model and effort choices, environment variables, or organization settings can change it. Effort controls how much reasoning the model does. For the cheaper everyday model, /model sonnet selects Sonnet 5.5 on a direct Anthropic connection in Claude Code 2.1.284 or later, also at medium effort by default.

Haiku 5.5 is available in Claude Code on a direct Anthropic connection from version 2.1.293. The model configuration docs map /model haiku to Haiku 5.5 there. The subagent docs also allow haiku or a full model ID in a custom subagent's model field; use claude-haiku-5-5 to pin the version. Provider aliases and organization settings still apply, so a third-party provider's haiku alias can point to Haiku 4.5.

In practice you upgrade because interruptions cost more than the higher fee. Check whether you are exhausting a session allowance, a weekly allowance, or a model-specific allowance before buying more capacity. The distinction matters more than a vague estimate of how many hours you code.

The limits that actually decide your bill

Inside a subscription's included usage, Claude Code does not bill you per task. You buy an allowance, and the question is how fast you spend it.

Four things shape that allowance:

It is shared across your Claude tools. Claude Code, the web, desktop, and mobile apps, Cowork, and Duet where available draw on the same plan limits.

Since October 21, 2026, Claude Duet's research preview has added another use of that allowance for invited US Pro and Max users. Its background briefs and check-ins count too; it is not yet available on Team or Enterprise. Budget for that work alongside coding, rather than assuming a separate Duet allowance.

Included usage resets on a rolling five-hour session window, with weekly limits on top. Pro, Max, Team, and legacy seat-based Enterprise users can see their allowance in Settings > Usage or Claude Code's /usage. You can have session capacity left and still reach the weekly limit. Max and Team support specify a fixed weekly reset time assigned to your account. Anthropic does not publish a fixed message count or a guaranteed coding-hours budget; task length, model, effort, and features change consumption. The usage-limit explanation distinguishes this allowance from a model's context window.

Fable access depends on your tier. Max, Team Premium, and legacy seat-based Enterprise Premium include Fable use up to 50% of the weekly limit, not 50% extra. Pro, Team Standard, and legacy seat-based Enterprise Standard use pay-as-you-go usage credits for Fable from the first token. Current usage-based Enterprise plans meter every model at API rates instead of supplying a seat allowance.

Heavy agent runs consume more allowance. When you point Claude Code at a big task and let it loop, reading files, running commands, and re-reading output, it processes more tokens than a short chat. Long conversations and repeated context can use up the week even when the number of prompts looks modest.

When you hit a limit, you can wait or pay to continue. These usage credits are a separate purchase from the included API grant.

1. 1

   Decide whether to keep going or wait

   With usage credits enabled, requests after an exhausted five-hour or weekly limit continue at standard API rates until your spending cap is reached or your balance runs out with auto-reload off. The regular reset schedule stays the same. On Claude Code 2.1.234 or later, /rate-limit-options can wait and resume the interrupted task after the reset. Waiting avoids extra usage charges; the monthly API grant cannot pay for those charges.

2. 2

   Stay strictly on your plan

   Keep usage credits disabled, log in with only your subscription credentials, and make sure ANTHROPIC_API_KEY is not set in that shell. Check /status to confirm the active credential. Then wait for the included allowance to reset when you reach a limit.

3. 3

   Watch your usage with /usage

   Run /usage inside Claude Code to see your current plan bars, activity statistics, and usage breakdown before you start a big agent run. It helps distinguish a recurring capacity problem from an unusually expensive task.

If you repeatedly hit the wall on Pro, assess Max 5x. If you exhaust that tier too, compare Max 20x with capped pay-as-you-go usage. A higher session multiplier does not remove the weekly limit.

Subscription vs API pay-as-you-go

The subscription is not the only way to run Claude Code. You can point it at the Anthropic API and pay per token, with no monthly API subscription required. Whether that is cheaper depends entirely on how much you run. Interactive Claude Code remains separately metered even when your Console organization has a monthly API credit grant.

Since October 7, 2026, Max and Team include monthly API credits, in the amounts shown in the plan table. The same announcement launched Haiku 5.5 and reduced Sonnet 5.5 cache-read prices. The credits reduce the cost of apps and agents you build; they add no interactive coding allowance.

The credits cover the Claude API, including Message Batches, the Claude Console Playground, Managed Agents, and the Claude Agent SDK. They do not cover interactive Claude Code in the terminal, IDE, desktop, or web, or extra usage in Claude, Claude Code, or Cowork. Free, Pro, and Enterprise are not eligible for this monthly grant.

You can claim after seven days on an active eligible plan. A Max subscriber uses Settings > Billing; a Team Owner or Primary Owner uses Organization settings > Billing. In the API credits section, choose Link organization and connect a Console organization where you hold an Owner, Admin, or Billing role. Credits arrive each billing cycle, monthly on annual plans, expire at that cycle's end, and are spent before purchased credits. Everyone with an API key in the linked organization draws from that balance.

For non-interactive work, self-run claude -p and Agent SDK calls can use the grant when authenticated with an API key from that linked organization. Signing in with your Claude plan instead uses the plan's limits. Runs started by the Claude Code GitHub Action, an IDE extension, or the desktop app are excluded even with -p, as the credit rules specify.

These are the current API rates from Anthropic's live rate card. The latest lineup is Fable 5.1, Opus 5.5, Sonnet 5.5, and Haiku 5.5. Haiku 4.5 remains available as a legacy model. MTok means one million tokens.

Released September 28, 2026, Claude Sonnet 5.5 remains the latest Sonnet release. Its fresh-input and output rates match Sonnet 5, now listed under Legacy models. Its cache reads cost less than Sonnet 5's.

Claude Code 2.1.278 made eligible auto-mode classifier checks server-side by default for Enterprise and Claude API accounts, with platform-dependent rollout on Bedrock, Google Cloud's Agent Platform, Microsoft Foundry, and gateways. A server-side check carries no classifier charge. /status shows Auto mode server: Enabled; if those checks cannot reach the session, Claude Code warns about fallback to its own classifier requests, which are billed as before. This removes classifier overhead from eligible metered sessions without changing subscription or token prices.

Claude Code 2.1.257 made Fable 5.1 the default Fable alias on a direct Anthropic connection. Its cache reads cost less than legacy Fable 5's, while fresh input and output rates are unchanged. The Fable 5.1 explainer covers the migration.

Two levers cut the API bill. Prompt caching lets you reuse a large context such as a codebase or long system prompt. Current cache-read rates per million tokens are Fable 5.1 at $0.25, Opus 5.5 at $0.20, Sonnet 5.5 at $0.10, Haiku 5.5 at $0.01 for prompts up to 100k tokens or $0.05 above that threshold, and Haiku 4.5 at $0.10. These are read charges only; original input, cache writes, and output are separate. Batch processing saves 50% on input and output for workloads that do not need an instant answer.

So when does API beat a subscription? For interactive Claude Code, it wins when your metered total stays below the flat fee. For your own apps and agents, API access is the route, and eligible Max or Team credits can offset that spend. Caching reduces repeated-input costs, not output charges. A Max subscription makes steady daily use more predictable, but it still has session and weekly limits.

If you already pay for Max and spend on eligible API work, claiming the grant can replace some purchased API credits within that billing cycle. Your subscription fee stays the same, and interactive Claude Code extra usage still costs separately. Treat the grant as an app-development budget, not a refund or more coding capacity.

Teams and enterprise

For groups, pricing moves to per-seat, and both Team seat types include Claude Code. Premium buys more capacity, not access to the coding agent.

Team Standard carries 1.25x Pro usage per session; Premium carries 6.25x. Included Claude and Claude Code usage limits are per member, not pooled across the team, and cannot be transferred between members. Owners can enable usage credits at standard API rates after a member exhausts the included allowance. You can mix seat types, so the economical setup is Standard for most people and Premium for members who repeatedly exhaust Standard. The separate monthly API grant is pooled and capped as shown in the plan table; adding seats beyond that cap does not increase the grant. That API pool cannot top up a member's interactive Claude Code allowance.

Current Enterprise plans charge a seat fee plus usage at API rates. The seat fee covers Claude, Claude Code, and Cowork access but includes no token allowance. Spending limits still stop usage, and self-serve plans also stop when their purchased credit balance runs out. Older Chat, Chat + Claude Code, and Standard or Premium Enterprise seats remain legacy tiers until renewal. Enterprise adds SCIM for user provisioning, audit logs, custom data retention, compliance APIs, and organization or user spend limits.

Which plan is right for you

For a solo builder, start on Pro and let recurring interruptions inform an upgrade. Before paying more, check /usage and the model and effort you are using. A rare spike may suit capped usage credits; frequent session-limit interruptions make Max worth assessing. Max 20x is for a workload that still outgrows Max 5x, with weekly limits retained.

For a small team, choose Team Standard when centralized billing and administration matter, then assign Premium where members need the extra per-session capacity. Do not buy Team expecting unused coding allowance to move from one colleague to another. For bursty or automated workloads, compare the API meter with the subscription fee and claim an eligible grant if you already have Max or Team.

For most people reading this, the answer is Pro for a month, then watch your /usage. If you are curious how the cost stacks up against the obvious alternatives, the math against Cursor is laid out in Claude Code vs Cursor, and if budget is the hard constraint, Gemini CLI vs Claude Code covers the comparison.

Is the Pro plan worth it for Claude Code?

For most solo builders, yes. It includes both Claude Code and the Claude apps, and it is the cheapest individual subscription with Code access. Capacity varies by task, model, and effort; assess Max when recurring interruptions cost more than the higher fee. Current prices are in the plan table above.

How do I run Claude Code for free, or as cheaply as possible?

Claude Code is not included on the Free subscription. Pro is the cheapest individual subscription route, with a lower advertised monthly equivalent if you pay annually upfront. API access has no subscription fee but bills usage, and there is no standing Max discount beyond occasional official promotions.

Is Claude Code Max worth it?

If you frequently exhaust Pro, the higher per-session allowance can justify Max. Max 20x is worth assessing when Max 5x still runs dry. Both retain weekly limits. Their monthly API grants fund eligible apps and agents, not interactive Claude Code or extra usage.

Was Claude Code removed from Pro?

No. Claude Code is included in Pro as well as Max. Anthropic's current pricing and support documentation explicitly cover using Claude Code on either plan.

Why is Claude Code so expensive?

A heavy agent run reads files, runs commands, and loops, consuming more tokens than a short chat. Included usage can run out, and usage credits or API authentication can add metered charges. Watch /usage for consumption and /status for the active credential before changing plans.