Claude Pro vs Claude Max: Complete Comparison and Report on Price, Usage Limits, Fable Access, and API Credits
Claude Pro and Claude Max are the two individual Anthropic subscriptions that sit between the free tier and the $200 Ultra plan. Pro costs $20 a month billed monthly, or $200 a year, which the official pricing page presents as $17 a month. Max costs $100 a month and is sold only as a monthly subscription. Since October 30, 2026, when Anthropic renamed the old Max 20x tier to Ultra, "Max" refers to a single plan, the one that gives five times the Pro allowance. Both plans include Claude Code, Research, memory, Projects, Opus 5.5, Sonnet 5.5 and Haiku 5.5, and both draw chat, desktop, mobile and Claude Code activity from one shared pool of usage.
The verdict is direct once the numbers are lined up. Max gives 5x the Pro usage every five hours and every week for 5x the monthly Pro price, so one unit of usage costs $20 on both plans, and the annual Pro plan is cheaper per unit at about $16.67. What the extra $80 a month buys, beyond volume, is Claude Fable included for up to 50% of the weekly limit (Pro users pay for Fable through usage credits at API rates), a $100 monthly credit for the Claude API that cannot be spent in the Claude apps or in interactive Claude Code, and the early-access and priority-access items listed on the pricing page. Pro is the better choice for anyone who stays inside one allowance most weeks and uses Fable only occasionally. Max is the better choice for people who hit the Pro limits regularly, who want Fable 5.2 as a daily model, or who would spend around $100 a month on the Claude API anyway.
All prices, limits and plan terms below were verified on Anthropic's official pricing page, help center articles and API documentation on November 12, 2026. Prices exclude applicable tax, and Anthropic states that mobile app prices may differ by platform.
··········
CLAUDE PRO AND CLAUDE MAX IN NOVEMBER 2026: WHAT EACH PLAN IS.
Pro is Anthropic's entry paid plan for individuals. The pricing page describes it as "More usage than Free" and treats it as the 1x baseline for every other individual plan, while the usage FAQ on the same page says Pro gives at least five times the Free usage per five-hour session. Pro adds Claude Code, Research, Projects without the Free plan's cap of five, the Opus model line, Claude Science, task handoff and scheduling, Claude in Chrome and Microsoft 365, and the Claude Data beta with Sheets and Whiteboards that reached all paid plans on November 10.
Max is the plan for heavier individual use. Its help center article defines it as "five times the Pro plan's usage every five hours and every week" and lists Claude Code in the same subscription, longer multi-step tasks that continue in the background, priority access to new models and products, and monthly API credits. The Max line used to include two tiers, Max 5x and Max 20x. On October 30, 2026 Anthropic moved Max 20x subscribers to the new Ultra plan at the same $200 price, removed the five-hour window for verified Ultra accounts and left Max as the single $100 tier. Our comparison of Claude Max and Claude Ultra covers the step above Max.
The table below summarizes the identity of the two plans as Anthropic publishes them today.
........
Item | Claude Pro | Claude Max |
Monthly price | $20 | $100 |
Annual option | $200 a year ($17/month shown) | Not offered |
Usage relative to Pro | 1x (baseline) | 5x every 5 hours and every week |
Fable models | Through usage credits | Included up to 50% of weekly limit |
Monthly API credits | None | $100 for the Claude Platform |
Claude Code | Included | Included |
Usage credits beyond limits | Available | Available |
........
··········
CLAUDE PRO VS CLAUDE MAX PRICE, BILLING OPTIONS AND TAXES.
The price gap depends on how Pro is billed. Against monthly Pro at $20, Max costs $80 more each month. Against annual Pro, whose $200 charge is billed up front, the effective monthly gap is about $83.33, and over a year the two plans cost $200 and $1,200 respectively. A monthly Pro subscriber pays $240 over twelve months, so the annual commitment saves $40 a year for users who are certain they will stay on Pro.
Max has no annual discount. The Max help article states that the plan is available only as a monthly subscription, and the pricing page shows the same monthly-only billing for Ultra. Both pages note that web prices may differ from those charged through the iOS and Android apps, that pricing varies by region and local currency for Pro, and that the listed prices exclude tax.
........
Billing scenario | Claude Pro | Claude Max | Difference |
One month, monthly billing | $20 | $100 | $80 |
Twelve months, monthly billing | $240 | $1,200 | $960 |
Twelve months, Pro annual | $200 | $1,200 | $1,000 |
Effective monthly, Pro annual | $16.67 | $100 | $83.33 |
........
··········
CLAUDE PRO AND MAX USAGE LIMITS: FIVE-HOUR WINDOW, WEEKLY CAP AND THE 5X MULTIPLIER.
Both plans use the same two-layer limit system. A session limit starts with the first message and resets in full five hours later, and a weekly limit applies across all models and resets at a fixed time assigned to each account. Max adds a third element, a separate weekly limit for Fable. The help center article on usage limits explains that consumption depends on conversation length and complexity, the features in use, the model and the effort level, with higher effort consuming more. Anthropic publishes no fixed message counts or token budgets for either plan, and it reserves the right to add weekly, monthly, model or feature caps at its discretion.
The 5x multiplier applies to both layers. A Max subscriber gets five times the Pro allowance inside each five-hour window and five times the Pro allowance across the week. Claude.ai on the web, the desktop app, the mobile apps and Claude Code all draw from this same pool, so a long coding session in the terminal reduces what is left for chat in the browser.
........
Limit element | Claude Pro | Claude Max |
Five-hour session limit | Yes, 1x | Yes, 5x Pro |
Weekly limit, all models | Yes, 1x | Yes, 5x Pro |
Separate weekly Fable limit | No (Fable billed to credits) | Yes, up to 50% of weekly |
Shared across apps and Claude Code | Yes | Yes |
Published message or token counts | None | None |
Usage credits after the limit | Standard API rates | Standard API rates |
Where to check | Settings > Usage | Settings > Usage |
........
Claude Code received a specific increase earlier in the year. On 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 reduction for Pro and Max accounts, after a compute agreement that gave the company the capacity of the Colossus 1 data center. That change applied to both plans equally, so it did not alter the 5x ratio between them.
··········
WHAT ONE UNIT OF CLAUDE USAGE COSTS ON PRO AND ON MAX.
Because Anthropic expresses every individual plan as a multiple of Pro, the price of one Pro-sized allowance can be computed directly from the published figures. The arithmetic below was run in Python on the prices listed on November 12, 2026.
........
Plan | Price per month | Allowance (Pro = 1x) | Price per 1x allowance |
Claude Pro, monthly | $20 | 1x | $20.00 |
Claude Pro, annual | $16.67 effective | 1x | $16.67 |
Claude Max | $100 | 5x (5-hour and weekly) | $20.00 |
Claude Ultra | $200 | 10x weekly | $20.00 |
........
The result removes a common assumption about the higher plan. Max carries no volume discount over monthly Pro, and the annual Pro plan is the cheapest way to buy Claude usage per unit on any individual tier. A Pro user who needs more capacity has two published routes. One is to upgrade, which multiplies the allowance by five. The other is to stay on Pro and enable usage credits, which bill the excess at standard API rates and can be prepaid at a discount through usage bundles.
The comparison by unit price leaves out the items that only Max includes. The Fable allowance, the $100 API credit and the early-access and priority-access entries on the pricing page are the real differentiators, and their value depends on the type of user.
··········
CLAUDE FABLE 5.2 ON PRO AND MAX: USAGE CREDITS VERSUS THE 50% WEEKLY SHARE.
Fable is Anthropic's most capable generally available model family, and access to it is the sharpest difference between the two plans. The help center article on Fable models states that on Pro, Fable models "are not part of your plan's usage limits" and are paid with usage credits from the first request. On Max and Ultra, Fable is included and subscribers can spend up to 50% of their weekly limits on it at no extra cost. Fable consumes those limits faster than other models, and the article specifies that the 50% share sits inside the weekly limit and does not add to it.
Fable 5.2 launched on November 9, 2026 and is available on every paid plan, together with Fable 5.1 and Fable 5. In the Claude apps and Claude Code its default effort is medium. In Claude Code, Fable models are never the default on any plan and have to be selected explicitly, and interactive sessions show a consent prompt before Fable usage is billed to credits. Our report on the Claude Fable 5.2 launch covers the model's benchmarks and API pricing.
A temporary exception applies this week. Under the Fable 5.2 launch offers, Pro subscribers whose paid billing started before November 9, 2026 at 10:00 am PT have Fable included in their weekly limits, up to 50%, from November 9 through November 16, 2026 at 11:59 pm PT, plus one Fable limit reset that expires at the same time. After November 16, Fable on Pro returns to usage credits. Eligible Max subscribers receive one to three Fable limit resets depending on how long they have subscribed (one under six months, two between six and twelve months, three over twelve months), valid until March 8, 2027.
........
Fable item | Claude Pro | Claude Max |
Fable 5.2, 5.1 and 5 available | Yes | Yes |
Counted against plan limits | No, billed to usage credits | Yes, up to 50% of weekly limit |
Cost beyond the included share | Standard API rates | Standard API rates |
Launch offer, Nov 9 to Nov 16, 2026 | Included up to 50% of weekly limit, 1 reset | Not applicable |
Launch offer resets | 1, expires Nov 16, 2026 | 1 to 3, expire Mar 8, 2027 |
Default in Claude Code | Never default | Never default |
........
··········
CLAUDE MODELS ON EACH PLAN AND THE API RATES BEHIND USAGE CREDITS.
The pricing page lists Opus, Sonnet and Haiku on Pro, with Fable through usage credits, and Opus, Sonnet, Haiku and Fable on Max. In practice the current models are Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.2, all with a 1M-token context window and 128K maximum output on the API according to Anthropic's models overview. The help center states that on paid plans the newest models support up to 1M tokens of context in the Claude apps, and it does not distinguish Pro from Max on context length.
Usage credits are billed at the standard API rates, so the API price list is also the price list for any Pro or Max usage beyond the plan. The table reproduces the official rates for the current models.
........
Model | Input per 1M tokens | Output per 1M tokens | Cache hit per 1M | 5-minute cache write per 1M |
Claude Fable 5.2 | $8 | $40 | $0.20 | $10 |
Claude Opus 5.5 | $4 | $20 | $0.20 | $5 |
Claude Sonnet 5.5 | $2 | $10 | $0.10 | $2.50 |
Claude Haiku 5.5 (prompt up to 100K) | $0.10 | $0.50 | $0.01 | $0.125 |
Claude Haiku 5.5 (prompt over 100K) | $0.50 | $2.50 | $0.05 | $0.625 |
........
The Batch API halves input and output prices, but batch pricing applies to API usage only, which matters for the Max API credit discussed below and not for chat or Claude Code overages.
··········
[[ADPLUS_300x250_1]]
CLAUDE CODE ON PRO AND MAX: DEFAULT MODEL, SHARED LIMITS AND EFFORT.
Claude Code is included on both plans with the same model line-up. According to the Claude Code model configuration documentation, the default model on Pro, Max, Team, Enterprise and the Anthropic API is now Opus 5.5. Before Claude Code version 2.1.280 the defaults differed by plan, with Sonnet 5 on Pro and Team Standard seats and Opus 5 on Max, so the current release gives Pro subscribers the same default model as Max subscribers. The difference between the plans in the terminal is therefore the size of the allowance and Fable access, with identical model choices for Opus, Sonnet and Haiku.
Claude Code and the Claude apps share one pool on both plans. When a Pro or Max subscriber reaches the five-hour limit mid-response, Claude Code may finish the current step under a capped wrap-up allowance that counts toward the weekly limit. Opus 5.5, Sonnet 5.5 and Haiku 5.5 support the low, medium, high, xhigh and max effort levels and default to medium, and the help center notes that higher effort consumes more of the allowance. Selecting a lower effort level is one of the few published ways to stretch a Pro allowance without paying for credits.
The $100 API credit on Max does not extend the Claude Code allowance. The help center article on Claude Code with paid plans states that monthly API credits included with Max or Ultra cannot be used in Claude Code or in the Claude apps.
··········
FEATURES THAT DIFFER BETWEEN CLAUDE PRO AND CLAUDE MAX.
Outside usage volume, Fable and API credits, the two plans share almost every product feature. Memory, connectors, Artifacts, file creation, code execution, Research, Projects, Claude in Chrome and Microsoft 365, Claude Science and the Sheets and Whiteboards beta appear on both. The Personal Finance beta, launched on November 5, 2026, is available on Pro, Max and Ultra for US accounts. The Claude Duet research preview, announced on October 21, is open to Pro, Max and Ultra in the US through a waitlist.
Two items are listed differently on Anthropic's own pages. The pricing page shows "early access to advanced features" and "priority access at high-traffic times" as Max additions over Pro, while the Pro help center article lists priority access during high-traffic periods and early access to new features among Pro's benefits. The pricing page comparison table is the more recent and more specific of the two, but the overlap means neither item can be treated as a guaranteed exclusive of Max.
........
Feature | Claude Pro | Claude Max |
Claude Code | Yes | Yes |
Research | Yes | Yes |
Projects | Yes | Yes |
Memory and connectors | Yes | Yes |
Claude Data, Sheets and Whiteboards (beta) | Yes | Yes |
Personal Finance (beta, US) | Yes | Yes |
Claude Duet (research preview, US waitlist) | Yes, 3 limit resets | Yes, 1 limit reset |
Early access to advanced features | Listed on Pro help page only | Listed on pricing page |
Priority access at high-traffic times | Listed on Pro help page only | Listed on pricing page |
Claude Cowork | Not listed for individual plans | Not listed for individual plans |
........
The Duet line has an unusual detail. The Duet help article grants Pro subscribers three limit resets and Max and Ultra subscribers one each, all expiring on December 15, 2026. Duet work, including background tasks such as the morning brief, draws from the plan's five-hour and weekly limits at the same rate as chat, so the extra resets partly offset Pro's smaller allowance for Duet users.
··········
USAGE CREDITS AND USAGE BUNDLES ON BOTH PLANS.
Usage credits are the published way to keep working after a limit on either plan. Subscribers enable them in Settings > Usage, can set a monthly spend limit or leave it unlimited, and can turn on auto-reload on the web. The balance covers Claude on web, desktop and mobile, Claude Code, Cowork and third-party products connected to the Claude account. Credits bought before December 7, 2026 do not expire, while credits bought on or after that date expire 12 months after purchase, subject to country rules.
Usage bundles are prepaid credit packs sold at a discount on Pro, Max, Ultra and Team. Bundle credits apply only after the plan's included usage is exhausted, purchases are non-refundable, and individual subscribers can buy up to $2,000 of discounted bundles per month, with anything beyond billed at standard rates.
........
Usage bundle | Price paid | Discount on standard credit rates |
$50 of credits | $45 | 10% |
$250 of credits | $200 | 20% |
$1,000 of credits | $700 | 30% |
........
The bundles change the Pro-versus-Max arithmetic for users with predictable overflow. A Pro subscriber who buys a $250 bundle pays $200 for usage worth $250 at API rates, which is more than the $80 monthly gap to Max but can last several months for occasional Fable work, provided it is used within the expiry window that applies after December 7.
··········
THE $100 MONTHLY API CREDIT ON CLAUDE MAX.
Max includes $100 a month of credits for the Claude Platform, a benefit that Anthropic introduced for Max and Team on October 7, 2026. Pro is not eligible. Subscribers claim the credit in Settings > Billing by linking a Claude Console organization in which they hold the Owner, Admin or Billing role, and the plan must have been active for seven days before the first claim.
The credit covers the Claude API (Messages and Message Batches), the Console Playground, Claude Managed Agents and the Claude Agent SDK. It does not cover interactive Claude Code, extra usage in Claude, Claude Code or Cowork, or Claude models accessed through Amazon Bedrock, Google Cloud Vertex AI or Microsoft Foundry. Credits refresh each billing cycle, unused credits expire at the end of the cycle, and everyone with an API key in the linked organization shares the balance. Upgrading from Max to Ultra brings prorated credits at once and $200 a cycle afterwards.
For a developer, the credit offsets most of the price gap. Python arithmetic on the official rates shows that $100 buys 10 million input tokens and 3 million output tokens on Opus 5.5 at standard prices, and the same volume costs $50 through the Batch API, so the credit stretches to roughly twice that volume for batch workloads. A developer who already spends $80 to $100 a month on the Claude API pays close to nothing for the move from Pro to Max, as long as the spend fits the credit's scope and is used before each cycle ends.
··········
COST SCENARIOS: WHEN PRO PLUS CREDITS BEATS A MAX SUBSCRIPTION.
The scenarios below compare the $80 monthly gap between monthly Pro and Max with the cost of buying the same extra work as usage credits on Pro. All arithmetic was run in Python on the official API rates of November 12, 2026. The token volumes are assumptions chosen for illustration, and Anthropic does not publish how many such sessions fit inside the Max Fable share, so the scenarios measure the price of the work, not a guaranteed Max allowance.
Scenario A is a Fable 5.2 coding session of 2.0 million input tokens, 1.8 million of them read from cache at $0.20 per million and 0.2 million written to the five-minute cache at $10 per million, plus 150,000 output tokens at $40 per million. One session costs $8.36 on usage credits. Scenario B is an Opus 5.5 analysis of a 300,000-token document with a 5,000-token answer, without caching, which costs $1.30 per request. Scenario C is the API credit check from the previous section.
........
Scenario (on Pro usage credits) | Standard rates | With 20% bundle | With 30% bundle |
5 Fable 5.2 sessions a month | $41.80 | $33.44 | $29.26 |
10 Fable 5.2 sessions a month | $83.60 | $66.88 | $58.52 |
15 Fable 5.2 sessions a month | $125.40 | $100.32 | $87.78 |
20 Fable 5.2 sessions a month | $167.20 | $133.76 | $117.04 |
60 Opus 5.5 long-document requests | $78.00 | $62.40 | $54.60 |
Break-even vs $80 Max gap (Fable sessions) | 9.6 | 12.0 | 13.7 |
........
The break-even point sits at about ten Fable sessions of this size a month at standard rates, and between twelve and fourteen sessions when credits are bought in the $250 or $1,000 bundles, which require paying in advance. Below that volume, Pro with credits costs less than Max. Above it, Max is cheaper for the Fable work alone, before counting the fivefold allowance for Opus, Sonnet and Haiku and the $100 API credit. The Opus 5.5 scenario shows a second route to the same conclusion, since a Pro user who repeatedly exceeds the allowance with long-document work reaches the $80 gap within about sixty requests a month at standard rates.
··········
UPGRADING FROM CLAUDE PRO TO CLAUDE MAX AND BEYOND.
The Max help article describes the upgrade paths. An annual Pro subscriber who moves to Max receives the remaining Pro balance as a credit for future charges when it exceeds the Max price, and the billing address must match the previous subscription, otherwise Pro has to be cancelled and allowed to end before subscribing to Max. Subscribers billed through Google Play pay the full new price, and the unused value of the old plan becomes extra days that push the renewal date back. Moving from Max to Ultra is charged as a prorated amount for the rest of the billing cycle. The article does not describe a downgrade process from Max to Pro.
New limits apply as soon as an upgrade completes, according to the help center article on usage limits. Users who are unsure can therefore start on Pro, watch the Settings > Usage page for a few weeks, and upgrade on the day the limits start getting in the way, without losing the current billing period. Our guide to Claude plans and usage limits covers the Free, Team and Enterprise tiers that sit around these two plans.
··········
WHAT ANTHROPIC DOES NOT PUBLISH ABOUT CLAUDE PRO AND MAX ALLOWANCES.
Several facts a buyer would want are absent from Anthropic's pages. There is no published number of messages, tokens or hours for either plan, only the 5x ratio between them, and the pricing page and help center both say usage varies with model, effort, features and conversation length. Anthropic does not state how quickly Fable consumes the weekly limit compared with Opus 5.5, so the practical size of Max's 50% Fable share is unknown. The rate limits behind the May 6 Claude Code increase appeared only in an image, without figures in the text.
The plan pages also disagree on two benefits. Early access and priority access appear as Max additions on the pricing page and as Pro benefits in the Pro help article. Claude Cowork does not appear in the individual-plan tables on the pricing page, while the usage-credit and bundle articles say that credits cover Cowork. No independent evaluation organization measures consumer plan allowances, so there are no third-party figures to set against Anthropic's descriptions.
··········
CLAUDE PRO OR CLAUDE MAX: THE CHOICE BY TYPE OF USER.
The decision follows from three verified facts: the same $20 price per unit of usage on monthly Pro and on Max, Fable on Pro billed entirely to credits after November 16, and a $100 API credit that only Max includes and that cannot be spent in the apps or in Claude Code.
........
Type of user | Recommended plan | Reason |
Daily chat and writing within limits | Claude Pro (annual) | Lowest price per unit at $16.67 a month |
Occasional Fable 5.2 use | Claude Pro plus credits | Under about 10 sessions a month costs less than $80 |
Fable 5.2 as the main model | Claude Max | Up to 50% of a 5x weekly limit included |
Claude Code most working days | Claude Max | 5x five-hour and weekly allowance, same default model |
Developer spending $80 to $100 on the API | Claude Max | $100 monthly API credit offsets the price gap |
Heavy but irregular overflow | Claude Pro plus bundles | Prepaid credits at up to 30% off, no monthly commitment |
Near the Max weekly limit every week | Claude Ultra | 10x weekly, no five-hour window when verified |
........
Most individual users who chat, write and code a few hours a week are served by Pro, preferably on annual billing, with usage credits enabled as a safety valve. Max pays for itself in three cases: when the Pro limits interrupt work on most days, when Fable 5.2 is the model of choice for long coding or reasoning sessions beyond roughly ten a month, or when the subscriber is also a developer whose regular Claude API bill is close to $100. For a full view of every tier with its prices, see our Claude pricing guide.
··········
FOLLOW US FOR MORE.
·····
DATA STUDIOS
·····
datastudios.org
