A claude.ai login draws the plan’s shared pool. That pool resets on a rolling five-hour window, and paid plans add a weekly cap. Web, desktop, mobile, and Claude Code share it; Anthropic publishes no message count. A Console login bills tokens at the API list on claude.com/pricing, including Opus 5.5 at $4 input and $20 output per million tokens. Amazon Bedrock, Google Cloud’s Agent Platform, and Microsoft Foundry bill the cloud account per token; these pages do not publish those clouds’ token prices. The Claude Code changelog now defaults Pro and Team Standard to Opus, matching Max, Team Premium, and Enterprise, which can spend the shared pool faster. Confirm the sign-in before a long session.
Current prices for every path that includes Claude Code
Claude Code is Anthropic’s agentic coding assistant. It reads a repository, edits files, runs commands, and works in the terminal, supported IDEs, the Claude desktop app, and the browser. A paid Claude plan includes it. Anthropic’s Plans & Pricing page, as of 27 September 2026, lists Pro at $17 per month with the annual subscription discount, billed as $200 up front, or $20 if billed monthly. The rest of the stickers are in the table below.
Annual subscriptions are available for Pro and Team. Max is billed monthly. Enterprise is annual only. Anthropic says annual plans are designed to cost less than monthly subscriptions. Team seats can be mixed. Anthropic does not publish a second monthly dollar figure for Max 20x.
The Enterprise pricing card is US$20 per seat per month billed annually, plus usage at API rates that scale with the model and task. Sales-assisted Enterprise can ask Anthropic for a tailored quote. Separately, the Claude Code costs doc says usage inside a Team or Enterprise seat allowance is not metered in dollars, and the organization’s spend report appears once usage credits are on. These pages do not say how the seat price, the API-rate usage, and the seat allowance combine on one bill.
The pricing page describes a discounted university-wide Education plan for an institution’s students, faculty, and staff. It does not publish a price. Usage limits apply on every plan. The prices shown exclude applicable tax. Anthropic states that price and plans are subject to change at its discretion.
The Claude Code changelog states that the default model on Pro and Team Standard changed from Sonnet to Opus, matching Max, Team Premium, and Enterprise. That product change can spend the shared allowance faster.
- Every plan’s usage resets on a rolling five-hour session window.
- Paid plans also have weekly limits.
- Web, desktop, mobile, and Claude Code share one pool.
- Anthropic publishes no fixed message count; how much you can do depends on conversation length, the model, and the features you use.
- These ratios are usage multiples, and Anthropic does not turn them into a message count: Pro is at least 5x Free per five-hour session; Max is 5x or 20x Pro; Team Standard is more than Pro; Premium is 5x Standard.
Official list prices from claude.com/pricing as of 27 September 2026. Usage limits apply. Prices exclude tax.
| Plan | List price | Billing cycle | Claude Code |
| — | — | — | — |
| Free | $0 | n/a | No |
| Pro | $17 per month with annual discount ($200 billed up front); $20 billed monthly | Monthly or annual | Yes |
| Max | From $100 per month; choose 5x or 20x the usage of Pro per five-hour session | Monthly | Yes |
| Team Standard | $20 per seat per month billed annually; $25 billed monthly | Monthly or annual | Yes |
| Team Premium | $100 per seat per month billed annually; $125 billed monthly | Monthly or annual | Yes |
| Enterprise | US$20 per seat per month billed annually, plus usage at API rates | Annual | Yes |
Plans & Pricing | Claude by Anthropic
Is Claude Code free?
The Free comparison row on Anthropic’s pricing page marks Claude Code as No. Chat on web, iOS, Android, and desktop remains on Free, along with web search, Artifacts, memory, and code execution in chat. Claude Code is included in all paid plans: Pro, Max, Team, and Enterprise. Terminal work and chats draw from one shared usage pool, so a long coding session and a long chat both consume the same allowance.
The Claude desktop app bundles Claude Code in its Code tab, so a separate CLI install is optional there. A paid subscription is required for that Code tab. Most surfaces need a Claude subscription or an Anthropic Console account. The terminal CLI, VS Code, and JetBrains also support third-party providers.
Pro and Max are one unified subscription for Claude on web, desktop, and mobile and for Claude Code in the terminal and in supported IDEs, including VS Code, Cursor and other VS Code forks, and JetBrains IDEs such as IntelliJ and PyCharm. Log in with the same Claude credentials you use on claude.ai. If Claude Code is already signed in through Claude Console pay-as-you-go, run /login inside Claude Code to switch to the subscription.
Use Claude Code with your Pro or Max plan
What a Console or cloud login costs
A Console login bills Claude Code by token at Anthropic’s API list rates. Anthropic’s pricing page on 27 September 2026 lists these rates per million tokens. Prompt-cache read and write prices sit on the same cards and reflect a five-minute TTL. For workloads that need to run in the US, US-only inference is available at 1.1x on input and output tokens. Fast mode for Opus 5.5 is priced separately at 2x standard.
Amazon Bedrock, Google Cloud’s Agent Platform, and Microsoft Foundry bill the cloud account per token and carry no Claude seat fee. Spend controls live in that cloud billing console. These pages do not publish those clouds’ token prices. Claude Code’s costs doc says token billing applies on Console or a cloud provider, and points subscription stickers to claude.com/pricing. The first Console authentication auto-creates a workspace named Claude Code for centralized tracking.
Official API list rates from claude.com/pricing as of 27 September 2026, for a Console login. MTok means per million tokens. Prompt-cache prices reflect a 5-minute TTL.
| Model | Input | Output | Cache read | Cache write |
| — | — | — | — | — |
| Opus 5.5 | $4 / MTok | $20 / MTok | $0.20 / MTok | $5 / MTok |
| Sonnet 5 | $2 / MTok | $10 / MTok | $0.20 / MTok | $2.50 / MTok |
| Haiku 4.5 | $1 / MTok | $5 / MTok | $0.10 / MTok | $1.25 / MTok |
| Fable 5.1 | $10 / MTok | $50 / MTok | $0.25 / MTok | $12.50 / MTok |
Plans & Pricing | Claude by Anthropic
How billing follows the login
Cost depends on how you authenticate. A claude.ai Pro, Max, Team, or Enterprise login draws the plan or seat allowance. Console or API authentication bills per token at Anthropic’s API list. Amazon Bedrock, Google Cloud’s Agent Platform, and Microsoft Foundry bill the cloud account per token. If an organization mixes sign-in methods, each developer is metered by the method they authenticated with.
If ANTHROPIC_API_KEY is set, Claude Code authenticates with that API key, which produces API charges on Pro, Max, Team, or Enterprise accounts. On first use, a set key skips the login prompt and asks you to approve the key.
Use Claude Code with your Pro or Max plan
Check claude.com/pricing for the current list. Confirm which credential Claude Code is using: for a Pro or Max allowance, sign in with those Claude credentials and leave ANTHROPIC_API_KEY unset; run /login if you are still on Console pay-as-you-go. Every plan resets on a rolling five-hour session window, and paid plans add a weekly cap. After you hit either window, wait for it to reset, move to a higher plan, or on a paid plan turn on usage credits at standard API rates. Switching to Console API credits is a separate offer and needs an explicit yes. Subscribers should read the plan bars in /usage. On Console, the session total in /usage is a local estimate; the invoice is the Console Usage page.