How much Claude Code dey cost?

How much Claude Code cost?

Claude Code cost $20 every month for Claude Pro plan, $100 or $200 every month for the two Max tiers, or e fit charge you per token if you pay with API credits from Claude Console. Free version no dey. Subscription dey give you usage allowance wey Claude Code share with Claude chat. API credits dey buy tokens, and dem dey bill am at the published rate per million tokens. Monthly limit no dey unless you set one. Every figure for this page na as of 2 August 2026.

Which option cheaper depend on how many hours you run am every day and whether you need hard limit for spending. Numbers for both options dey below, together with one worked estimate for one developer wey dey work alone.

Claude Code free dey?

No. Claude Code no get free tier. Anthropic setup documentation talk am clearly: Claude Code need Pro, Max, Team, Enterprise, or Console account, and free Claude.ai plan no include access to Claude Code. You fit install the command line tool without paying anything, and claude --version go print version number, but login step no accept free account, so no session go start. If you already dey pay for Pro and you wan know wetin e cover before you install anything, wetin Pro subscription dey give you for Claude Code answer am plan by plan.

Two things dey make people think say free tier dey. The first one na free Claude.ai plan, wey cover chat for web and apps, and na there e stop. wetin free Claude plan really include explain where that boundary dey. The second one na small trial credit wey Anthropic give new Claude Console account to test API. Na one-time credit be that, no be monthly allowance. When e finish, requests go stop.

The cheapest real entry point na Pro at $20 every month, or $17 every month if you pay for one year upfront.

Subscription prices as of August 2026

[
  {
    "plan": "Free",
    "price_usd": 0
  },
  {
    "plan": "Pro",
    "price_usd": 20
  },
  {
    "plan": "Pro, billed annually",
    "price_usd": 17
  },
  {
    "plan": "Max 5x",
    "price_usd": 100
  },
  {
    "plan": "Max 20x",
    "price_usd": 200
  },
  {
    "plan": "Team, standard seat",
    "price_usd": 25
  },
  {
    "plan": "Team, premium seat",
    "price_usd": 125
  }
]

Pro at $20 per month na the entry plan wey include Claude Code. Max get two sizes. Max 5x at $100 per month give five times the per-session usage of Pro, while Max 20x at $200 give twenty times. Team dey billed per person: $25 per month for standard seat and $125 for premium seat. Both cheaper when you bill annually. Enterprise na seat price plus usage wey dem charge at API rates, and sales team go give you the quote. If you dey buy for group instead of yourself, wetin each Team and Enterprise seat really include dey compare those seat prices with the cost of paying per token.

Wetin subscription give you na usage allowance, e no be token budget wey you fit inspect. Usage dey reset on rolling five hour window. Max tiers add two weekly windows: one wey apply across all models, and one wey apply only to Sonnet models. Claude chat and Claude Cowork share this allowance, so time wey you spend for chat in the morning go reduce wetin remain for coding afternoon. The complete Claude tier chooser compare the plans with each other in detail.

API credits per token dey cost how much

If you sign Claude Code in with Claude Console account instead, no monthly fee dey. You go buy credits, and the tool go spend dem per token. MTok below mean one million tokens.

[
  {
    "label": "Opus 5",
    "input_per_mtok_usd": 5,
    "output_per_mtok_usd": 25,
    "cache_read_per_mtok_usd": "0.50"
  },
  {
    "label": "Sonnet 5, to 31 Aug 2026",
    "input_per_mtok_usd": 2,
    "output_per_mtok_usd": 10,
    "cache_read_per_mtok_usd": "0.20"
  },
  {
    "label": "Sonnet 5, from 1 Sep 2026",
    "input_per_mtok_usd": 3,
    "output_per_mtok_usd": 15,
    "cache_read_per_mtok_usd": "0.30"
  },
  {
    "label": "Haiku 4.5",
    "input_per_mtok_usd": 1,
    "output_per_mtok_usd": 5,
    "cache_read_per_mtok_usd": "0.10"
  }
]

Opus 5 cost $5 per million input tokens and $25 per million output tokens. Sonnet 5 dey use introductory pricing of $2 and $10 until 31 August 2026. After that, e go change to $3 and $15 on 1 September 2026. Haiku 4.5 na the cheapest option at $1 and $5.

The main rate no be the full bill. Claude Code dey send your complete conversation again with every request. So, most of your input tokens dey come as cache reads. Dem cost one-tenth of the base input price: $0.50 per MTok for Opus 5. Writing to the cache cost pass base input. E cost 1.25 times the input rate for the five minute cache and 2 times for the one hour cache. Extended thinking dey billed as output tokens. Na the expensive column above dey apply. So, if you leave model for high effort setting, e fit cost real money for every turn.

Model choice na the biggest control wey you get. Choosing which Claude model to run fit change the cost of the same work by five times.

Which billing path dey cost less for how you dey use am

Subscription fit work well for interactive work. You sit for terminal, code for some hours, then stop. The bill fixed, and the limit wey you go reach na usage window, no be invoice wey you no expect.

API credits fit anything wey no be person dey type. Scheduled jobs, continuous integration runs, scripted agents, and per-user cost reporting all need token billing, because personal subscription allowance dey tied to one human account and e no report anything per project. API credits na the only option wey get real spend cap, because workspace spend limits for Claude Console fit stop spending when e reach the number wey you choose.

The full comparison of API billing against a Claude subscription explain where the crossover dey, so this page no go repeat the argument. Short answer be say: if one person dey use Claude Code for most working days, subscription dey win. Decision only dey close for light use or use wey dey happen in big bursts. If you still dey choose whether to use any tool, how Claude Code compares with the other coding agents put the subscription prices side by side.

A workable monthly estimate for one solo developer

Make we assume say one developer dey work alone, about three hours every day, 20 active days for one month, Sonnet na the default model, and Opus na for the hard parts. As of August 2026, Anthropic publish two figures for deployments wey dey bill by token: average cost na about $13 for each developer per active day, while 90% of users spend below $30 per active day. Multiply each one by 20 active days to get the monthly amount.

[
  {
    "label": "Pro subscription",
    "price_usd": 20
  },
  {
    "label": "Max 5x subscription",
    "price_usd": 100
  },
  {
    "label": "Max 20x subscription",
    "price_usd": 200
  },
  {
    "label": "API credits, published average",
    "price_usd": 260
  },
  {
    "label": "API credits, heavier tenth of users",
    "price_usd": 600
  }
]

For token billing, that developer go pay around $260 every month based on the published average. If dem fall inside the heavier 10%, e fit reach about $600. Pro at $20 and Max 20x at $200 both dey below that amount.

Read the difference well, because e no mean say the money free. Pro no give you $260 worth of tokens for $20. E give you an allowance, and three hours every day with Sonnet plus Opus for difficult parts fit reach the five-hour window on busy days. The correct way to understand this table be say Max 5x at $100 per month match this usage pattern without stopping you in the middle of a task, and e still cost less than the average token bill. The two Max tiers dey use the same models and get the same features, so to choose between the two Max prices depend on how often your workdays reach that window. If Pro limits dey stop you often, upgrading fit cost less than the tokens for that usage. Every figure here na only half of the calculation, because plan price no matter as much as the value e save you, and whether that monthly price go pay for itself at your own hourly rate na the calculation wey make sense to run before you choose a tier.

How to see wetin you really dey spend

Run /usage inside Claude Code session. For subscription, e go show plan usage bars plus breakdown of which subagents and MCP servers your recent usage go to. E go also flag any behaviour wey cause 10% or more of that usage. Press d or w to switch between the last 24 hours and the last 7 days.

For API billing, the Session block for top of /usage go print dollar amount:

Total cost:            $0.55
Total duration (API):  6m 20s
Total code changes:    0 lines added, 0 lines removed

The system calculate that amount locally from token counts using list rates. So e no know about promotional pricing or negotiated discount, and e fit differ from your invoice. The authoritative number na the usage page for Claude Console. Session totals also reset when /clear start new session. So small number after clear no mean say you get cheap day. How to read Claude Code token usage line by line explain wetin each counter dey count.

When you reach plan limit, /usage-credits go open billing settings. For there, you fit turn on usage credits and continue work after the allowance, with token rates. For Team or Enterprise plan, the same command go send request to whoever get billing access.

Wetin dey make bill high

Long session fit cost pass the work wey e look like, because every request dey carry the complete conversation. Four habits dey change the amount pass anything else:

• Run /clear between unrelated tasks. Old context dey send again for every message wey follow am, so if you put unrelated work for one session throughout afternoon, you go pay for the morning work again and again.
• Change models with /model. Sonnet fit handle most coding work, and e output rate dey far below Opus, na where thinking tokens dey land.
• Monitor the cache lifetime. If you return to session after long break, system go process your full context again at the base input price. The cache dey last one hour for subscription and five minutes for API key.
• Keep agent teams small. Every teammate dey run im own context window. When teammates dey run for plan mode, dem use roughly seven times the tokens of standard session.

How to keep Claude Code context window small na the same skill as keeping bill small. How much the agent writes matter just as much, because model wey rewrite complete file when three lines fit solve am go charge for every one of those lines at output rate. Na so e be for a skill wey make agent stick to the smallest change wey go work. If you run agents unattended, to cap wetin AI agent fit spend for VPS matter even more, because nobody dey watch the session while e dey burn tokens.

VPS dey run am cost extra?

Yes, and e go come as separate bill. If you run Claude Code for server instead of laptop, na the server you go pay for. Claude Code need 4 GB RAM or more, plus supported operating system like Ubuntu 20.04 or later, so small VPS dey enough. Install am the same way you go install am locally:

curl -fsSL https://claude.ai/install.sh | bash
claude --version

If installation work, e go print version string like 2.1.211 (Claude Code). If command report command not found, installer don put binary for ~/.local/bin, but your shell never pick am up: open new shell and try again. Then run claude for your project directory and log in through browser prompt, using the paid account wey you already get. Your plan follow the account, no be the machine, so VPS session dey use the same allowance as your laptop.

The reason to move go server no be cost. Na because server session dey continue even after you close your laptop. Run Claude Code for VPS inside tmux cover that setup, and the permissions to set before you leave agent running cover wetin you need lock down first. Add server cost on top of your plan, and wetin VPS really cost every month give the current range.

FAQ

Claude Code free to use?

No. Claude Code need paid Claude plan (Pro, Max, Team or Enterprise), or Claude Console account wey get API credits. Anthropic setup documentation talk say free Claude.ai plan no include Claude Code access. So login step go fail for free account, even though installation dey work. New Console accounts dey receive small one-time trial credit to test API. This one no be recurring free tier.

Claude Pro subscription include Claude Code?

Yes. Pro cost $20 per month, or $17 per month when dem bill am yearly. E include Claude Code together with Claude chat. Dem share the usage allowance between both, and e reset for rolling five hour window. If you dey code for several hours every day, you go reach this window. Na why Max tiers wey cost $100 and $200 per month dey available.

Subscription or API billing, which one cheaper for daily coding?

For one person wey dey code most working days, subscription almost always cheaper. Anthropic published average for token-billed deployments na about $13 per developer per active day. This one na roughly $260 across 20 active days, compared with $100 per month for Max 5x. API billing better when you need hard spend cap and per-user cost reporting. Personal subscription no provide either of these.

How I fit check how much Claude Code dey cost me?

Run /usage for session. For subscription, e show how much of your plan window you don use and which parts of the session use am. For API billing, Session block show dollar total wey e calculate locally with list prices. This amount fit differ from your real invoice, so use usage page for Claude Console as the correct amount. Session total go reset when you run /clear. Check am before you clear am.

Wetin go happen when I reach my Claude Code usage limit?

Claude Code go tell you which window you reach and when e go reset. Changing models with /model no restore access, because session and weekly windows dey shared across models. But e fit let you continue work after model-specific message such as an Opus limit. Run /usage-credits to turn on usage credits and continue after you don use your allowance, with token rates. If credits no dey, wait until the window reset.