HomeBez kategoriiClaude Code Pricing 2026: Plans, API Costs and Limits

Claude Code Pricing 2026: Plans, API Costs and Limits

Last updated: November 7, 2026 – Author: Ignacy Kwiecien, founder and editor-in-chief, DecodeTheFuture.org

Updated November 7, 2026

Anthropic renamed its top plans on October 30, 2026: Max 5x is now simply Max ($100/month) and Max 20x is now Ultra ($200/month). Ultra gives 10x Pro usage per week and, for verified accounts, no five-hour limit. This guide now reflects the current plan names, Team seat prices, the 2026 Claude Code limit changes and current API model prices. For the full Ultra breakdown, see our Claude Ultra plan explainer.

Claude Code pricing in 2026 has two different routes: a Claude Pro, Max or Ultra subscription with usage limits shared with the Claude app, or pay-as-you-go Anthropic API billing. Pro is $20 per month ($17 billed annually), Max is $100 per month with 5x Pro usage, and Ultra (formerly Max 20x) is $200 per month with 10x Pro usage per week. API cost depends on the model and tokens used. Max and Ultra include monthly API credits, but Anthropic says those credits cannot be used in Claude Code.

Pro $20Max $100Ultra $200API metered5-hour and weekly limits

Claude Code pricing at a glance

Claude Code is a terminal coding agent. You can authenticate it with a paid Claude plan, with an Anthropic Console account that pays for API usage, or through an enterprise platform such as Amazon Bedrock or Google Vertex AI. The login path matters because it determines how usage is billed.

Route Public price What you are paying for Best fit
Claude Pro $20/month
$200/year ($17/month)
Claude app plus Claude Code access, with five-hour and weekly limits shared between both Light or moderate solo coding
Claude Max (formerly Max 5x) $100/month 5x Pro usage, with five-hour and weekly limits; includes $100/month in API credits that cannot be used in Claude Code Frequent daily coding
Claude Ultra (formerly Max 20x) $200/month, monthly billing only 10x Pro usage per week; no five-hour limit for verified accounts; includes $200/month in API credits that cannot be used in Claude Code Heavy agent loops and large repositories
Anthropic API Metered by model and tokens Input, output and optional cache charges on the Console route Irregular use, automation and cost metering
Team Standard seat $20 annual / $25 monthly; Premium seat $100 annual / $125 monthly 2 to 150 users, Claude Code included, admin controls, up to $500/month in pooled API credits Small and mid-size teams
Enterprise $20/seat/month billed annually (20-seat minimum) plus usage at API rates Central billing, governance and usage-based pricing Organizations that need governance

Prices above are US list prices before tax at the time of this update. Regional checkout, taxes, annual billing and product eligibility can change the amount shown in your account. Use Anthropic’s current pricing page as the final price check.

What does a Claude subscription include?

With Pro, Max or Ultra, Claude Code is connected to the same Claude account used by the Claude app. Anthropic describes the limits as shared: activity in Claude and Claude Code counts against the same allowance. That makes a subscription predictable at the monthly invoice level, but it does not mean that every coding session has an unlimited token budget.

Important billing distinction

A Pro, Max or Ultra subscription is not a wallet of Anthropic API dollars for Claude Code. Max and Ultra do include $100 and $200 per month in Claude API credits, but Anthropic says those credits cannot be used in Claude Code or the Claude apps. If you switch Claude Code to an Anthropic Console account or set an ANTHROPIC_API_KEY, usage is billed at API rates instead of counting against your plan. Once you hit a plan limit, you can keep working with optional usage credits, which are also billed at standard API rates.

Pro: the entry point

Pro costs $20 per month, or $17 per month billed annually ($200 up front). It has a five-hour session limit and a weekly limit, both shared with the Claude app. Anthropic does not publish a fixed number of Claude Code prompts per window. Older estimates of roughly 10-40 prompts per five-hour window predate the May 2026 doubling of Claude Code’s five-hour limits and should be read as historical. Repository size, prompt length, conversation history, attachments, model choice and tool use can move the limit substantially.

Max and Ultra: more capacity, not unlimited capacity

Max costs $100 per month and gives 5x Pro usage, with both a five-hour limit and a weekly limit. Ultra, the plan formerly called Max 20x, costs $200 per month on monthly billing only and gives 10x Pro usage per week. Verified Ultra accounts have no five-hour limit; unverified accounts get a temporary one until verification completes. The weekly limit still applies, and both Max and Ultra have a separate weekly limit for Fable. The multiplier describes capacity relative to Pro; it is not a promise that five or ten times every specific workflow will fit. Anthropic also notes that how fast you reach a limit depends on the model you use, so Opus-heavy work runs out sooner than work on lighter models. Our Ultra plan explainer covers why the old “20x” name no longer matches the weekly allowance.

How Claude Code limits changed in 2026

  • May 6, 2026: Anthropic doubled Claude Code’s five-hour rate limits for Pro, Max, Team and seat-based Enterprise plans, and removed the peak-hours limit reduction for Pro and Max.
  • May 13 – September 13, 2026: a promotion raised weekly Claude Code limits by 50%. Five-hour limits were not affected.
  • From September 14, 2026: weekly Claude Code limits are 25% higher than before the promotion on Pro, Max, Ultra, Team and seat-based Enterprise plans.
  • October 30, 2026: Max 5x became Max and Max 20x became Ultra, and the five-hour limit was removed for verified Ultra accounts. The same day, fast mode for Opus 4.8 was deprecated; it will be removed on November 30, 2026. On subscription plans, fast mode is billed only from usage credits, not from plan limits.

Claude Code API pricing: how the metered route works

The API route charges for model tokens rather than a flat monthly seat. The Anthropic Platform pricing page is the source of truth for the active model IDs, input prices, output prices and prompt-cache rates.

API cost component Why Claude Code can increase it How to control it
Input tokens The agent reads files, tool results and conversation context Keep tasks scoped; use focused repository instructions
Output tokens Long plans, patches and explanations add generated tokens Ask for concise changes and inspect diffs between steps
Prompt-cache writes Large reusable context may be written to cache Use caching deliberately and check the current cache TTL
Prompt-cache reads Repeated context can be cheaper than a fresh input Keep related work in one session when practical
Search or external tools Some tool calls can have their own charge or add tokens Separate discovery from high-volume implementation loops
Model (API list price) Input per 1M tokens Output per 1M tokens Cache hit per 1M tokens
Claude Opus 5.5 (Claude Code’s default model) $4 $20 $0.20
Claude Sonnet 5.5 $2 $10 $0.10
Claude Haiku 5.5 (prompts up to 100K tokens) $0.10 $0.50 $0.01
Claude Haiku 5.5 (prompts over 100K tokens) $0.50 $2.50 $0.05

Prices are from Anthropic’s pricing page as of November 7, 2026. Opus 5.5 is the default Claude Code model on Pro, Max, Ultra, Team, Enterprise and the API; fast mode on Opus 5.5 costs $8/$40 per million tokens.

Do not estimate API cost from the number of terminal prompts alone. One prompt that loads a large repository can be more expensive than several short questions. The correct unit is the model’s billed input and output, not the visible count of messages.

Subscription versus API: a simple break-even model

Use this formula for a first estimate, then compare it with your real Console usage:

Formula
monthly API cost =
  (input tokens / 1,000,000 x input price)
  + (output tokens / 1,000,000 x output price)
  + cache and tool charges

For example, at Sonnet 5.5’s list price of $2 per million input tokens and $10 per million output tokens, a month with 2 million input tokens and 300,000 output tokens is about $7.00 before cache or tool charges. On Opus 5.5 ($4/$20), the same volume is about $14.00. Both are below a $20 Pro subscription on raw token math. Add several large repository contexts and longer output loops, and the same workflow can cross the subscription price quickly.

Practical rule

Choose the API when your usage is irregular, automatable or easy to meter. Choose Pro, Max or Ultra when you use Claude Code often enough that a predictable invoice and shared Claude access are worth more than exact per-token control.

What about Claude Team and Enterprise billing?

The Team plan covers 2 to 150 users. On Anthropic’s pricing page, a Standard seat costs $20 per seat per month billed annually or $25 billed monthly, and a Premium seat costs $100 per seat per month billed annually or $125 billed monthly. Anthropic describes a Premium seat as 5x the usage of a Standard seat. Both seat types include Claude Code, and the plan includes up to $500 per month in pooled API credits. Prices are US list prices before tax; your region can show different currency and pricing.

Enterprise is priced as a seat plus usage: US$20 per seat per month billed annually, with a 20-seat minimum, plus usage at API rates. On Team and seat-based Enterprise plans, each seat’s allowance resets on a rolling five-hour window and a weekly window and is shared with Claude chat. Confirm the seat type and the usage-credits setting in the workspace admin console before budgeting for a team.

Which Claude Code plan should you choose?

Your situation Starting point Why
Occasional fixes or small repositories Pro Lowest predictable subscription cost and enough capacity for light work
Daily work with repeated refactors Max ($100) 5x Pro headroom without moving to a metered application budget
Large repositories and long agent sessions Ultra ($200) 10x Pro weekly usage and no five-hour limit once verified, but still subject to a weekly limit
CI jobs, scripts or sporadic bursts API Pay for actual model usage and set an explicit spend cap
Multiple developers with central governance Team or Enterprise Seat management, admin controls and usage reporting

Claude Code pricing versus Cursor and Codex

This article is intentionally not a head-to-head comparison. The existing Cursor vs Claude Code 2026 guide owns the direct tool comparison. Here the useful distinction is billing mechanics: Claude Code combines subscription access with Anthropic’s API route, while other coding tools can use different request pools, seats or model providers.

For a wider model-cost view, see our LLM inference cost comparison. Do not copy its model table into a Claude Code budget without checking which model IDs the coding product actually exposes.

Key takeaways

  • Pro is $20/month, Max is $100/month and Ultra (formerly Max 20x) is $200/month; all three share limits between Claude and Claude Code.
  • Verified Ultra accounts have only a weekly limit; Pro and Max keep both five-hour and weekly limits.
  • API billing is separate and depends on model tokens, caching and tool activity.
  • Anthropic’s prompt ranges are estimates, not fixed quotas; repository context changes the economics.
  • Team seats cost $20-$25 (Standard) or $100-$125 (Premium) per month for 2 to 150 users; Enterprise is $20 per seat plus usage at API rates.
  • Recheck the official Anthropic pages before purchase because prices, model IDs and limits change.

Frequently Asked Questions

Is Claude Code free?

The CLI can be installed, but useful access requires a paid Claude Pro, Max or Ultra plan, a Team or Enterprise seat, Anthropic Console billing, or an enterprise cloud provider. Anthropic’s pricing page lists Claude Code as not included in the Free plan.

Is Claude Code cheaper with a subscription or an API key?

It depends on token volume and workflow shape. Subscription plans are easier to budget for frequent interactive use; API billing can be cheaper for irregular or small workloads and gives you direct metering.

Do Claude Pro and Claude Code have separate limits?

No. Anthropic says Claude and Claude Code usage share the same plan limits. Pro and Max have a five-hour session limit and a weekly limit; verified Ultra accounts have only the weekly limit. The practical capacity depends on message length, context, files, model choice and tool use.

Does Claude Code include API credits?

Not in a way Claude Code can use. Max and Ultra include $100 and $200 per month in Claude API credits, but Anthropic says those credits cannot be used in Claude Code or the Claude apps. If you use an API key, the tokens are billed under the active API account and price sheet.

What happened to Claude Max 20x?

On October 30, 2026, Anthropic renamed Max 20x to Ultra and Max 5x to Max. Prices did not change: Max is $100 per month and Ultra is $200 per month. Ultra gives 10x Pro usage per week, and verified Ultra accounts have no five-hour limit.

How much does Claude Team cost?

Team covers 2 to 150 users. A Standard seat is $20 per month billed annually or $25 monthly; a Premium seat, with 5x the usage of a Standard seat, is $100 annually or $125 monthly, before tax. Both include Claude Code.

RELATED ARTICLES

LEAVE A REPLY

Please enter your comment!
Please enter your name here

- Advertisment -

Most Popular

Recent Comments