Skip to main content

AnthropicAI Assistant

Claude

Anthropic's work assistant: chat, Projects, research, documents and slides, connectors to company tools, hand-off tasks, and @Claude in Slack on Free, Pro, Max, Team, and Enterprise plans.

The 30-Second Take

Pick Claude if your teams will hand it whole tasks across documents, Slack and connected apps.

That hand-off work is what people praise most, and it is also what uses up a seat's limits fastest. Yale licenses Claude Enterprise to faculty and staff with a separate HIPAA environment for patient data.

Best fit
Teams that write, analyze, and coordinate across documents, Slack, and connected apps.
Main constraint
Each paid seat has a 5-hour and a weekly usage limit, and hand-off tasks use them faster than chat.
First move
Put one department on Team Standard seats and read the usage analytics before buying more.

Why

Watch the plan: Team seats cap each person with a 5-hour and a weekly limit, while Enterprise bills every token at API rates. Read a pilot's usage analytics before you pick.

Practitioner sentiment, leaders and teams+44 Positive

27 practitioners, 101 sources reviewed, Updated Nov 1, 2026How this is measured

Read the GILD Review
Save a product your organization uses.Keep up with Claude without adding it to your stack.Visit Website
Operating stack
AI Knowledge Work

Comparisons

How does Claude compare with the alternatives?

See all 5 comparisons

  1. Claude vs ChatGPT

    • AI Assistants

    Seat prices match, so let the work decide. Give both the three tasks your teams repeat every week, on the seat type you would buy, then compare the output and how soon each seat reaches its limit.

  2. Claude vs Gemini

    • AI Assistants

    Start with the Gemini your Workspace plan includes in Gmail, Docs, and Meet, and add Claude seats for teams that hand off multi-step work or also live in Microsoft 365 and Slack. Claude reaches Google files through connectors and a Docs, Sheets, and Slides add-on in beta.

  3. Claude vs Microsoft Copilot

    • AI Assistants

    On Microsoft 365, start with Copilot, which is grounded in your Microsoft data and already offers Anthropic's models. Add Claude where teams need its own app, Cowork, and @Claude in Slack; its add-ins also work inside Excel, PowerPoint, and Word.

Operator Insights

What are business and product leaders learning from rolling Claude out across their teams?

See all 10 operator accounts

Buying and deployment

What business and product leaders should verify before rolling it out across teams.

Sourced from Anthropic documentation

How you buy it

US$25

Per seat per month for Team Standard, or US$20 billed annually. Pro is US$20 a month.

Source 1, Anthropic, Plans & pricing: Plans & pricingSource 2, Anthropic, Team plan: Team planSource 5, Anthropic, Monthly API credits for Max and Team plans: Monthly API credits for Max and Team plansSource 3, Anthropic, Enterprise plan: Enterprise plan

Details

Monthly prices: Pro US$20, or US$17 a month billed annually as US$200 up front; Max from US$100. Max 5x includes US$100 a month in Claude Platform API credits and Max 20x US$200. Team, for 2 to 150 seats, costs US$25 a Standard seat or US$20 annually, and US$125 a Premium seat or US$100 annually, with up to US$500 a month in pooled API credits. The API credits pay for the Claude API and for the Agent SDK or claude -p run with a linked Console API key, never for interactive Claude Code. Enterprise is US$20 billed annually plus all usage at API rates.

Cost drivers

5-hour limit

Plus a weekly limit shared by chat, Cowork, and Claude Code. Credits continue at API rates.

Source 1, Anthropic, Plans & pricing: Plans & pricingSource 6, Anthropic, How do usage and length limits work?: How do usage and length limits work?Source 2, Anthropic, Team plan: Team planSource 7, Anthropic, Claude Fable models on your plan: Claude Fable models on your planSource 3, Anthropic, Enterprise plan: Enterprise plan

Details

Usage. Pro, Max, and Team seats have a 5-hour session limit and a weekly limit shared by chat, Cowork, and Claude Code, and a Team Premium seat gets 5x a Standard seat. Usage credits continue past the limits at API rates, and Fable models run on credits for Pro and Team Standard seats. Usage-based Enterprise seats have no usage limits, and all their usage bills at API rates.

Where it runs

Web, desktop, mobile

Plus Chrome, Excel, PowerPoint, and Word on paid plans, and Slack in beta.

Source 8, claude.com, Download Claude: Download ClaudeSource 1, Anthropic, Plans & pricing: Plans & pricingSource 9, support.claude.com, Use Claude in Google Docs, Sheets, and Slides: Use Claude in Google Docs, Sheets, and SlidesSource 10, Anthropic, What is Claude Tag?: What is Claude Tag?Source 11, claude.com, Introducing Claude Duet: Introducing Claude DuetSource 12, support.claude.com, Get started with Claude Duet: Get started with Claude Duet

Details

Web, desktop for macOS and Windows (Linux in beta), and iOS and Android. Paid plans add Claude in Chrome and Claude for Excel, PowerPoint, and Word. In beta: Claude for Outlook, the Google Docs, Sheets, and Slides add-on, and @Claude in Slack on Team and Enterprise. Claude Duet is a separate app where Claude acts on connected accounts and asks before it sends, pays, signs, or deletes. It has been in pre-release on a waitlist for US Pro and Max subscribers since October 21, 2026, and is not yet offered on Team or Enterprise.

What to verify

20 seats

The self-serve Enterprise minimum. SCIM, audit logs, and custom retention are Enterprise only.

Source 4, Anthropic, Model training (consumer plans): Model training (consumer plans)Source 1, Anthropic, Plans & pricing: Plans & pricingSource 3, Anthropic, Enterprise plan: Enterprise planSource 13, support.claude.com, What certifications has Anthropic obtained?: What certifications has Anthropic obtained?

Details

Which data terms apply: Free, Pro, and Max chats are used for training when the user allows it, while Team and Enterprise content is not by default. SCIM, audit logs, the Compliance API, and custom retention are Enterprise only; self-serve Enterprise needs 20 seats and sales-assisted 50. For its commercial products, Claude for Work and the API, Anthropic lists ISO 27001:2022, ISO/IEC 42001:2023, SOC 2 Type I and Type II, and a HIPAA-ready configuration with a BAA available, which the pricing page offers on Enterprise only.

Signals

What is the current evidence saying about Claude?

A current read of what people and companies using Claude for work report. Every finding links to its sources.

Whose sources count

Practitioner sentiment, leaders and teams

+44 Positive net

of 28 sources that take a side, each author counted once. Small sample.

How this figure is made

Current sentiment

Positive, mixed and negative sources

The score counts them as 14 positive, 11 mixed and 2 negative votes, one per author.

The 30 sources that speak for a team or company lean positive. All 7 that discuss hand-off tasks praise them, connectors draw praise in 5 of 8 accounts, and 6 support enabling staff before a rollout. Data and admin controls come up in seven accounts, and 2 raise them as a problem, including a company of about 110 users whose accounts Anthropic suspended at once without warning.

Run a pilot on your team's recurring tasks to measure how often answers are used without rework.

Evidence field. All 101 sources, one mark each.

Arrange by

Each mark is one source: positive, mixed, negative, no opinion stated, or documentation (not counted).

By sentiment, Leaders and teams rolling it out. Every source GILD read, grouped as the field draws it.
GroupSourcesPositiveMixedNegativeNo opinion stated and not counted in the scoreDocumentation, checked and not counted
Positive15150000
Mixed11011000
Negative200200
Not counted200020
Documentation18000018
Outside this group5326151020
Every source GILD read, by the quarter it was published, as the field's time axis counts it. Documentation is dated by the day it was checked.
QuarterSources
Q1 20251
Q2 20253
Q3 20255
Q4 20259
Q1 202615
Q2 202624
Q3 202622
Q4 20264
Documentation, checked18

By sentiment

18 documentation pages check facts and are not counted; 53 sit outside this group.

Hover a mark for a quick look, click it or press Enter to open its source.Tap a mark to open its source, then step to its neighbors with Previous and Next.

What practitioners praise

Top praise of 2. Each mark is a source on it; filled marks praise it.

  1. Hand-off TasksAll 7 sources on it praise it

Most common concerns

Top concern of 3. Each mark is a source on it; filled marks raise it.

  1. Missing FeaturesAll 6 sources on it raise it as a concern
All 101 sources behind these findings
  1. Context Engineering Is Knowing When to Start OverMike Taylor, head of evals at Every (Every, media and software company whose team works in Claude and ChatGPT daily), Newsletter, Every, published Oct 16, 2026
  2. Comment on: The gap in shared understanding of LLM capability is wideninggranfalloon, Hacker News commenter (self-described: Individual legal and compliance consulting plus personal research), Hacker News comment, published Oct 12, 2026
  3. Plans & pricingAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  4. What is the Team plan?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  5. What is the Enterprise plan?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  6. How am I billed for my Enterprise plan?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  7. How do usage and length limits work?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  8. Usage limit best practicesAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  9. Claude Fable models on your planAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  10. What are projects?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  11. Use connectors to extend Claude's capabilitiesAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  12. Use Claude Cowork on Team and Enterprise plansAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  13. Use Claude Cowork safelyAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  14. What is Claude Tag?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  15. Configure custom data retention controls for Enterprise plansAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  16. Is my data used for model training? (consumer plans)Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  17. Is my data used for model training? (commercial products)Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  18. Can Claude produce images?Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  19. Claude Cowork and chat are now one ClaudeAnthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  20. Claude for Microsoft 365Anthropic (Official Claude documentation), Product documentation, checked Oct 12, 2026
  21. Comment on: The gap in shared understanding of LLM capability is wideningnasmorn, Hacker News commenter (self-described: Own monthly bookkeeping), Hacker News comment, published Oct 11, 2026
  22. Comment on: Dots: Always-on agentshungryhobbit, Hacker News commenter (self-described: Engineering team using Claude Tag in Slack for review routines), Hacker News comment, published Oct 1, 2026
  23. Now Available: Claude Enterprise Licenses at YaleAI at Yale, Yale University (self-described: University-managed Claude Enterprise for faculty and staff beside Microsoft 365 Copilot), University IT announcement, ai.yale.edu, published Sep 15, 2026
  24. Comment on: Ask HN: What default model do you use and why?joelboersma, Hacker News commenter (self-described: Own work on the Team plan; Codex free tier for personal projects), Hacker News comment, published Sep 12, 2026
  25. Comment on: Muse - Meta's personal AI agentfnordpiglet, Hacker News commenter (self-described: Solo household textbook shopping), Hacker News comment, published Sep 9, 2026
  26. How I AI: How this PM uses Claude to handle 70% to 80% of his workdayDaniel Blum, product manager at Melio (Melio, using Claude and Cowork on company-licensed tools, interviewed on How I AI), Podcast episode, How I AI (Lenny's Podcast Network), published Aug 31, 2026
  27. Comment on: Ask HN: What is one simple thing LLMs are insanely bad at?znnajdla, Hacker News commenter (self-described: Solo planning documents and email editing), Hacker News comment, published Aug 26, 2026
  28. Comment on: Manus will return to operating as an independent companyghshephard, Hacker News commenter (self-described: Own work at a corporation with restrictive InfoSec), Hacker News comment, published Aug 17, 2026
  29. Comment on: Manus will return to operating as an independent companyprimitivesuave, Hacker News commenter (self-described: Internal-services team enabling nontechnical coworkers), Hacker News comment, published Aug 11, 2026
  30. Comment on: Manus will return to operating as an independent companyghshephard, Hacker News commenter (self-described: Own knowledge work across mail, calendar, Jira, and Google Docs), Hacker News comment, published Aug 11, 2026
  31. Comment on: Manus will return to operating as an independent companymichelb, Hacker News commenter at a learning and development company (self-described: Learning and development company with about 15 coworkers), Hacker News comment, published Aug 11, 2026
  32. Comment on: Manus will return to operating as an independent companyriddlemethat, Hacker News commenter (self-described: Own compliance and project management work), Hacker News comment, published Aug 11, 2026
  33. Comment on: How I use LLMs to learn complex topicsbrooke2k, Hacker News commenter (self-described: Solo textbook study), Hacker News comment, published Aug 10, 2026
  34. Comment on: Ask HN: AI rollout for 100 employees, mostly non-technical - what worked?mentorcliq, Hacker News commenter administering an AI rollout (self-described: An organization that rolled AI out over two years, from ChatGPT Enterprise and Team toward Claude), Hacker News comment, published Aug 7, 2026
  35. Ask HN: AI rollout for 100 employees, mostly non-technical - what worked?eeemmmooo, Hacker News commenter leading an AI rollout (self-described: Corporate team of about 100 employees, mostly ops, finance, marketing, and site managers), Hacker News comment, published Aug 6, 2026
  36. Comment on: Ask HN: Sub 5 person team - Claude team plan?anoop_kumar, Hacker News commenter and startup founder (self-described: Founder of a team of fewer than five people), Hacker News comment, published Jul 27, 2026
  37. Comment on: Claude Cookbookitchingsphynx, Hacker News commenter (self-described: Solo recipe Project), Hacker News comment, published Jul 24, 2026
  38. An opinionated guide to which AI to useEthan Mollick, Wharton professor and author of One Useful Thing (self-described: Hands-on test of Claude Cowork and ChatGPT Work on his own Gmail), Newsletter post, One Useful Thing, published Jul 23, 2026
  39. MCP tool calls never reach Filesystem-type servers after Desktop updatejedchase-blip, GitHub user (self-described: Local MCP servers in Claude Desktop on macOS), GitHub issue, claude-ai-mcp, published Jul 22, 2026
  40. How the founder of Morning Brew built a Claude content machine that never runs out of ideas and never sounds like slopAlex Lieberman, Managing Partner at Tenex (Tenex, running company content through a Claude workflow and an employee posting program), Podcast episode, How I AI (Lenny's Podcast Network), published Jul 20, 2026
  41. Comment on: Setting up your spare Mac for Claude Code to control, a step-by-step guideintromert, Hacker News commenter (self-described: Solo laptop backup and reinstall project), Hacker News comment, published Jul 19, 2026
  42. I let Claude read all my newsletters for meAI editor at Tom's Guide (self-described: Solo, own work), Independent hands-on review, Tom's Guide, published Jul 15, 2026
  43. Snowflake MCP connector fails in claude.ai chat but works in Claude Code on same accountaaronbruneau, GitHub user (self-described: Snowflake connector in the Claude app for macOS, chat and Code tabs), GitHub issue, claude-ai-mcp, published Jul 15, 2026
  44. Comment on: ChatGPT Workneil_s, Hacker News commenter (self-described: Solo use), Hacker News comment, published Jul 9, 2026
  45. Comment on: Ask HN: Sub 5 person team - Claude team plan?lalithaar, Hacker News commenter (self-described: A team of fewer than five people), Hacker News comment, published Jun 30, 2026
  46. How to roll out Claude for Enterprise to all employees and get 73x ROIJustin Grieshop, CBTS (self-described: CBTS, IT services company that also sells AI rollout services, after rolling Claude Enterprise out to all employees), Company blog post, CBTS, published Jun 20, 2026
  47. Comment on: Show HN: Veterinarian turned founder, AI lawn diagnosisktrnka, Hacker News commenter (self-described: Individual native-habitat restoration work for Green Seattle), Hacker News comment, published Jun 15, 2026
  48. Comment on: Show HN: Veterinarian turned founder, AI lawn diagnosisjmathai, Hacker News commenter (self-described: Solo home and lawn project), Hacker News comment, published Jun 15, 2026
  49. Account-scoped auth failure: Gmail and Calendar return "The caller does not have permission" since April 20OvercastJeff, GitHub user (self-described: Gmail and Google Calendar connectors on a company Google Workspace account, in Cowork and claude.ai), GitHub issue, claude-ai-mcp, published Jun 10, 2026
  50. Comment on: Tell HN: Dont use Claude Design, lost access to my projects after unsubscribingjrumbut, Hacker News commenter (self-described: Solo use), Hacker News comment, published May 14, 2026
  51. Comment on: The Siri for Families Apple Will Never Buildabustamam, Hacker News commenter (self-described: Solo family reference Project), Hacker News comment, published May 14, 2026
  52. Tell HN: Dont use Claude Design, lost access to my projects after unsubscribingpycassa, Hacker News commenter (self-described: Own Claude Design projects after a Max subscription ended), Hacker News post, published May 13, 2026
  53. Cowork Mondays: The Bet I'm Making in 2026Scott Monday, operator of Trinity Builder Solutions and kitchen & bath CRATE (self-described: Two construction and remodeling companies moving most of their office staff onto Claude Cowork), Newsletter, Substack, published May 9, 2026
  54. Comment on: Anthropic bans orgs without warningtaklimakan, Hacker News commenter (self-described: A new company account), Hacker News comment, published Apr 24, 2026
  55. Comment on: Google plans to invest up to US$40B in Anthropicjonlucc, Hacker News commenter (self-described: Institutional animal-care protocol review and policy work with colleagues), Hacker News comment, published Apr 24, 2026
  56. Making Claude Cowork actually usefulCharles Cook, VP Operations, PostHog (PostHog, running scheduled Cowork tasks across about fifty small teams' plans, Slack, and GitHub), Company blog post, PostHog, published Apr 24, 2026
  57. Comment on: Gmail label bridge on Claude Cowork just brokealeixin, Hacker News commenter (self-described: Solo daily email drafting), Hacker News comment, published Apr 23, 2026
  58. Cowork on 3P: How to Run Any LLM in Claude Cowork and Claude CodePaweł Huryn, The Product Compass (self-described: Solo, own work), Newsletter, The Product Compass, published Apr 23, 2026
  59. What Claude Design is actually good for (and why Figma isn't dead yet)Claire Vo, founder of ChatPRD (ChatPRD, building a landing page and a branded deck in Claude Design for How I AI), Podcast episode, How I AI (Lenny's Podcast Network), published Apr 22, 2026
  60. Anthropic bans orgs without warningalpinisme, Hacker News commenter at an agricultural technology company (self-described: Agricultural technology company with about 110 users on a Claude Team plan and a separate API account), Hacker News post, published Apr 21, 2026
  61. Comment on: ChatGPT for Excelvipipiccf, Hacker News commenter (self-described: Own work with Excel documents), Hacker News comment, published Apr 16, 2026
  62. Salesforce Hosted MCP: "Authorization with the MCP server failed" on production Enterprise orgbbgit123, GitHub user (self-described: Custom connector from Claude to a production Salesforce Enterprise org), GitHub issue, claude-ai-mcp, published Apr 16, 2026
  63. Comment on: Ask HN: Is Claude Getting Worse?the_inspector, Hacker News commenter (self-described: Solo use), Hacker News comment, published Apr 15, 2026
  64. Comment on: Claude Cowork found me a flat to rent in London in just 5 daysmikepapadim, Hacker News commenter (self-described: Solo London apartment search), Hacker News comment, published Apr 15, 2026
  65. Comment on: ChatGPT for Excellateforwork, Hacker News commenter (self-described: Solo presentation creation), Hacker News comment, published Apr 15, 2026
  66. Comment on: ChatGPT for Excelnsiemsen, Hacker News commenter working in finance (self-described: A finance company that added Claude Enterprise licenses), Hacker News comment, published Apr 15, 2026
  67. Comment on: I still prefer MCP over skillshasyimibhar, Hacker News commenter (self-described: Six-person team, about half non-engineers, using internal operations and Clickhouse data), Hacker News comment, published Apr 10, 2026
  68. Why We Switched to Claude from ChatGPTCraig Iskowitz, Founder and CEO, Ezra Group (Ezra Group, wealth management consultancy that made Claude its primary AI platform after buying a company-wide ChatGPT license), Company blog post, WealthTech Today (Ezra Group), published Apr 1, 2026
  69. Claude Dispatch and the Power of InterfacesEthan Mollick, One Useful Thing (self-described: Solo, own work), Newsletter, One Useful Thing, published Mar 31, 2026
  70. Is Claude Cowork an Agent Yet? Honest Review (2026)Pawel Jozefiak, Digital Thoughts (thoughts.jock.pl) (self-described: Solo, own work), Newsletter, Digital Thoughts, published Mar 24, 2026
  71. Comment on: MCP is dead; long live MCPspiderfarmer, Hacker News commenter (self-described: Own website content in a remote CMS), Hacker News comment, published Mar 15, 2026
  72. Comment on: Show HN: Claude Tuner - Monitor your Claude usage and find the right planxlos21, Hacker News commenter (self-described: Company in South Korea using Claude heavily, assigning individual Max plans by usage and skipping Team), Hacker News comment, published Mar 10, 2026
  73. How to do AI analysis you can actually trustCaitlin Sullivan, user-research consultant and former Head of UXR at Spotify Business (self-described: Same analysis run over 100 times across Claude, ChatGPT and Gemini), Newsletter guest post, Lenny's Newsletter, published Feb 17, 2026
  74. Comment on: Claude Opus 4.6blueblisters, Hacker News commenter (self-described: Solo stock-market research comparison), Hacker News comment, published Feb 6, 2026
  75. How this PM uses MCPs to automate his meeting prep, CRM updates, and customer feedback synthesisReid Robinson, Lead Product Manager at Zapier (Zapier, connecting Claude to Zapier's MCP server for CRM, meeting, and support workflows), Podcast episode, How I AI (Lenny's Podcast Network), published Feb 2, 2026
  76. Comment on: Putting Gemini to Work in Chromemogili1, Hacker News commenter (self-described: Solo use), Hacker News comment, published Jan 29, 2026
  77. What is Claude Cowork?Senior content specialist at Zapier (self-described: Solo, own work), Company blog post, Zapier, published Jan 23, 2026
  78. Lenny's Podcast: Zevi Arnovitz on building as a non-technical PMZevi Arnovitz, product manager at Meta (self-described: Personal projects), Podcast episode, Lenny's Podcast, published Jan 18, 2026
  79. Latent Space: Brex's AI strategyJames Reggio, CTO of Brex (self-described: Company procurement of chat assistants), Podcast episode, Latent Space, published Jan 17, 2026
  80. Comment on: First impressions of Claude Coworkvessenes, Hacker News commenter (self-described: Solo email and document task), Hacker News comment, published Jan 15, 2026
  81. Vibe Check: Claude Cowork Is Claude Code for the Rest of UsStaff writer at Every (self-described: Every team testing Cowork, including Dan Shipper and Kieran Klaassen), Company blog post, Every, published Jan 13, 2026
  82. Comment on: Cowork: Claude Code for the rest of your workhk__2, Hacker News commenter (self-described: Solo accommodation search), Hacker News comment, published Jan 12, 2026
  83. First impressions of Claude Cowork, Anthropic's general agentSimon Willison, simonwillison.net (self-described: Solo, own work), Personal blog post, Simon Willison's Weblog, published Jan 12, 2026
  84. Comment on: Claude in Chromesimonw, Hacker News commenter (self-described: Solo Cloudflare dashboard task), Hacker News comment, published Dec 23, 2025
  85. Comment on: Claude in Chromediwu1989, Hacker News commenter (self-described: Solo comparison of rental-car discount codes), Hacker News comment, published Dec 21, 2025
  86. Comment on: Gemini 3 Flash: Frontier intelligence built for speedleemoore, Hacker News commenter (self-described: Fortune 100 financial company, enterprise-wide evaluation and rollout), Hacker News comment, published Dec 17, 2025
  87. Comment on: The trust collapse: Infinite AI content is awfulvisarga, Hacker News commenter (self-described: Solo research), Hacker News comment, published Nov 6, 2025
  88. Vibe Check: Claude Skills Need a 'Share' ButtonKatie Parrott, staff writer at Every (Every, building editorial Skills for her own writing and for the rest of the team), Newsletter, Every, published Nov 3, 2025
  89. Comment on: Claude outageesperent, Hacker News commenter (self-described: Individual account holder), Hacker News comment, published Nov 1, 2025
  90. Comment on: Claude Memorylabrador, Hacker News commenter (self-described: Solo use), Hacker News comment, published Oct 23, 2025
  91. Claude Skills are awesome, maybe a bigger deal than MCPSimon Willison, simonwillison.net (self-described: Solo, own work), Personal blog post, Simon Willison's Weblog, published Oct 16, 2025
  92. Comment on: Kagi Newsstavros, Hacker News commenter (self-described: Solo use), Hacker News comment, published Oct 1, 2025
  93. Comment on: Claude's memory architecture is the opposite of ChatGPT'sqgin, Hacker News commenter (self-described: Solo use), Hacker News comment, published Sep 11, 2025
  94. Claude Memory: A Different PhilosophyFormer crypto researcher (self-described: Solo, own work), Personal blog post, shloked.com, published Sep 11, 2025
  95. Comment on: Claude Code weekly rate limitsthrowup238, Hacker News commenter (self-described: Museum volunteer researching rare books, plus personal and project questions), Hacker News comment, published Jul 28, 2025
  96. Lenny's Podcast: Dan Shipper on how Every works with AIDan Shipper, CEO and cofounder of Every (self-described: About 15-person media and software company), Podcast episode, Lenny's Podcast, published Jul 17, 2025
  97. Lenny's Podcast: Andrew Wilkinson's AI stackAndrew Wilkinson, co-founder of Tiny (self-described: Personal stack of a holding-company founder), Podcast episode, Lenny's Podcast, published Jul 3, 2025
  98. I've tried all the leading AI chatbots - here's why I keep going back to ClaudeWriter at Tom's Guide (self-described: Solo, own work), Independent hands-on review, Tom's Guide, published Jun 28, 2025
  99. Comment on: Google AI Ultrajohnisgood, Hacker News commenter (self-described: Solo use), Hacker News comment, published May 20, 2025
  100. Comment on: Show HN: I open-sourced my AI toy company that runs on ESP32 and OpenAI realtimemst, Hacker News commenter (self-described: Solo use), Hacker News comment, published Apr 23, 2025
  101. How I Use ClaudeFernando Borretti, borretti.me (self-described: Solo, own work), Personal blog post, borretti.me, published Jan 1, 2025

Jan 2025 to Oct 2026.

83 sources from practitioners, by month and sentiment.

The full readExplore all Claude Signals6 praise themes, 6 concerns, 5 points of agreement and 1 disagreement across all three groups, with all 101 sources open to inspect.

Where It Fits

Where does Claude belong in the stack?

Open Where It Fits: the Stack Field and the full map

Claude is Core across Assist and supports Find, Create, Analyze, and Automate.

Business area
Company-wide
Function
Knowledge Work
Operating stack
AI Knowledge Work
Primary role
AI Assistant & Work Agent
Claude beside the rest of its stackThe Stack Field, every product on every layer
FootprintClaude across the AI Knowledge Work stack. Bar height is its role on each layer.

  1. Assist

    1. AI assistant: Core

      AI assistant, core. This is Claude's primary role: chat on web, desktop, and mobile with Projects, memory, Skills, and connectors. It also runs hand-off tasks that keep going after the laptop closes, since Cowork merged into one Claude on Pro and Max from September 16, 2026.

  2. Find

    1. Web research: Supports

      Web research, supports. Web search is on every plan. Research, on paid plans, runs searches that build on each other across the web and connected apps such as Gmail and Google Docs, and returns answers with citations. A dedicated research engine still owns fast cited lookup at scale.

    2. Enterprise search: Supports

      Enterprise search, supports. Team and Enterprise include enterprise search, a dedicated project that searches Slack, Microsoft 365, and custom connectors with each person's own permissions. A company-wide index across every system stays with an enterprise search product.

  3. Create

    1. Documents: Supports

      Documents, supports. Claude drafts in chat and creates files, and Claude Docs and Slides export to PowerPoint or PDF. The Word add-in edits with tracked changes, and the Google Docs add-in (beta) edits in place under version history. The team's documents and wiki stay in those tools.

    2. Meeting notes: Touches

      Meeting notes, touches. Claude does not record or transcribe meetings; it works on transcripts and notes that reach it through uploads or connectors, so capture stays with a meeting notes tool.

  4. Analyze

    1. Data analysis: Supports

      Data analysis, supports. Code execution analyzes uploaded data and builds spreadsheets, and Claude for Excel builds and edits models with formulas. Claude Dashboards is in beta on paid plans. Governed metrics and reporting stay in the BI tool.

  5. Automate

    1. Automation: Supports

      Automation, supports. Claude schedules recurring tasks and works across connected apps. @Claude in Slack (Claude Tag, beta on Team and Enterprise) watches Slack channels and runs routines under its own identity. Deterministic multi-app workflows with triggers and retries stay in an automation platform.

Market mapThe layers Claude covers, and who else owns each one.

  1. General AI assistant

    CoreAssist

  2. Web research & cited answers

    SupportsFind

  3. Enterprise search & company knowledge

    SupportsFind

  4. Writing & documents

    SupportsCreate

  5. Data analysis & reporting

    SupportsAnalyze

2 more layers on the full map. Core, Supports and Touches describe coverage. They say nothing about quality.

What your subscription unlocks

What does signing in with a Claude plan give you that an Anthropic API key does not?

IncludedNot on this tierOpens the vendor's own words51 claims from 43 Anthropic pages

Which tier gets what

CapabilityPro (Pro)$20 a monthMax 5x (Max 5x)$100 a monthMax 20x (Max 20x)$200 a monthTeam (Standard seat) (Team (Standard seat))$25 a seat a monthTeam (Premium seat) (Team (Premium seat))$125 a seat a monthEnterprise (Enterprise)$20 a seat a month
Coding agents
Claude Code in the cloud1Claude Code in the cloudCloud sessions run Claude Code on Anthropic machines against a GitHub repo from the browser or phone and end with a pull request.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseCloud sessions are available on Pro, Max, and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats.Anthropic: Use Claude Code in the cloud2Claude Code in the cloudCloud sessions run Claude Code on Anthropic machines against a GitHub repo from the browser or phone and end with a pull request.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseWhen finished, Claude will automatically create a pull request with changes for you to review.A cloud session is a Claude Code session that runs on cloud infrastructure instead of on your machine, working with code from your GitHub repositories. You can start one from Claude Code on the web at claude.ai/code, from Claude MobileAnthropic: Use Claude Code in the cloudPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Code Review on PRs3Code Review on PRsA fleet of agents reviews GitHub pull requests and posts inline findings, billed through usage credits.LimitBilled outside the plan's included usageTiersTeam (Standard seat), Team (Premium seat), EnterpriseCode Review is available on Team and Enterprise plans.Code Review analyzes your GitHub pull requests and posts findings as inline comments on the lines of code where it found issues. A fleet of specialized agentsCode Review usage is billed separately through usage credits and doesn't count against your plan's included usage.Anthropic: Set up Code Review for Claude CodeBilled outside the plan's included usagePro: not includedMax 5x: not includedMax 20x: not includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude Security4Claude SecurityEnterprise teams get continuous vulnerability scanning across connected repositories, with patches proposed for review.LimitUp to eight repositories per projectTiersEnterpriseClaude Security is now available in public beta for users on Enterprise plans.continuously watches your codebases for security vulnerabilities, tracks each finding until it is fixed, and suggests targeted patches for human reviewconnect up to eight repositories from GitHub or other Git providersAnthropic: Use Claude SecurityUp to eight repositories per projectPro: not includedMax 5x: not includedMax 20x: not includedTeam (Standard seat): not includedTeam (Premium seat): not includedEnterprise: included
Deep research
Research mode5Research modeResearch runs chained web and connector searches on Opus 5.5 whatever model the chat uses.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseResearch is available for users with paid Claude plans (Pro, Max, Team, or Enterprise) using Claude on the web, Claude Desktop, or Claude Mobile.Research runs on Claude Opus 5.5 and falls back to Claude Opus 4.8 when needed.It runs multiple searches that build on each otherClaude researches the web and your connected toolsAnthropic: Use research on ClaudePro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude Science app6Claude Science appA desktop research app where Claude runs analysis code in a sandbox and pulls from scientific databases.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude Science is available in beta for Pro, Max, Team, and Enterprise plansClaude Science is Anthropic's desktop app for scientific research, currently in beta.Claude writes and runs code in a sandbox, reads the folders you give it, pulls data from scientific databases through connectorsAnthropic: Get started with Claude SciencePro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Enterprise search7Enterprise searchA dedicated project searches company sources such as Slack and Microsoft 365 in one place.TiersTeam (Standard seat), Team (Premium seat), EnterpriseEnterprise search capabilities are available for users on Team and Enterprise plans.Search across multiple data sources (Slack, Microsoft 365, and more) in one place.Enterprise search adds a dedicated project for searching across your organization's knowledge sourcesAnthropic: Use enterprise searchPro: not includedMax 5x: not includedMax 20x: not includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Agents and browsers
Claude in Chrome8Claude in ChromeA Chrome extension lets Claude read, click and navigate websites from the side panel, Cowork or Claude Code.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude in Chrome is available on paid plans (Pro, Max, Team, and Enterprise), in the Chrome side panel, Claude Cowork, and Claude Code.Claude in Chrome is a browser extension that allows Claude to read, click, and navigate websites alongside you.Anthropic: Get started with Claude in ChromePro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude Cowork9Claude CoworkCowork hands Claude multi-step office work that runs in the cloud and follows the account across devices.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude Cowork is available on paid plans (Pro, Max, Team, and Enterprise)Cowork runs your sessions remotely in the cloud, so your sessions and files live with your Claude accountAnthropic: Get started with Claude CoworkPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude Duet preview11Claude Duet previewA message-style assistant that works in connected apps and asks before it sends a message or pays; US waitlist only.TiersPro, Max 5x, Max 20xClaude Duet is available as a research preview to users on the Pro and Max plans in the US, through a waitlist.it works in the apps you've connected, lets you know when something needs you, and is designed to ask before it sends a message, makes a paymentAnthropic: Get started with Claude DuetPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): not includedTeam (Premium seat): not includedEnterprise: not included
Show all 4Show fewer
Scheduled tasks10Scheduled tasksRecurring tasks deliver briefings and reports on a schedule with the same connectors and skills as regular tasks.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseScheduled tasks are available on all paid plans (Pro, Max, Team, and Enterprise), in Claude Cowork and in the new Claude experience.Scheduled tasks have access to the same capabilities as regular Cowork tasks, including connected tools, skills, and installed plugins.delivering finished outputs like reports, briefings, and summaries every timeAnthropic: Schedule recurring tasks in Claude CoworkPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Apps and connectors
Custom MCP connectors12Custom MCP connectorsAny remote MCP server can be added as a connector in Claude, Cowork and Claude Desktop; Free accounts get just one.LimitFree plan capped at one custom connectorTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseCustom connectors using remote MCP are available on Claude, Cowork, and Claude Desktop for users on Free, Pro, Max, Team, and Enterprise plans. Free users are limited to one custom connector.Anthropic: Get started with custom connectors using remote MCPFree plan capped at one custom connectorPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Desktop extensions13Desktop extensionsOne-click packages install local MCP servers into Claude Desktop, on every plan, Free included.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseDesktop extensions provide a streamlined way to install and manage local MCP servers through single-click installable packages.Anthropic: Getting Started with Local MCP Servers on Claude Desktop14Desktop extensionsOne-click packages install local MCP servers into Claude Desktop, on every plan, Free included.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseDesktop extensions are available to all users on Claude Desktop.Anthropic: Use connectors to extend Claude's capabilitiesPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude Tag in Slack15Claude Tag in SlackTagging @Claude in a Slack channel hands it a task it works on under its own identity with the team's tools.TiersTeam (Standard seat), Team (Premium seat), EnterpriseClaude Tag is available on Team and Enterprise plans in beta. Claude Tag works in Slack today.tag @Claude into a conversation and it takes on real work, using your organization's tools and the shared context around it. Claude works under its own identityAnthropic: What is Claude Tag?Pro: not includedMax 5x: not includedMax 20x: not includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Documents and work apps
Claude for Excel16Claude for ExcelAn Excel add-in that reads open workbooks, debugs formulas and builds models inside Excel.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseAn Excel add-in that integrates Claude into your spreadsheet workflow, for Pro, Max, Team, and Enterprise plans.Ask questions about open workbooks, adjust assumptions while preserving formula relationships, debug errors, and build or populate modelsAnthropic: Use Claude for ExcelPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Claude in Google Docs17Claude in Google DocsA Claude sidebar in Docs, Sheets and Slides edits the open file in place.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude for Google Workspace is available in beta for Pro, Max, Team, and Enterprise plans.Claude for Google Workspace adds a Claude sidebar to Google Docs, Google Sheets, and Google Slides. Claude reads the file you have open and edits it directlyAnthropic: Use Claude in Google Docs, Sheets, and SlidesPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Unlimited projects19Unlimited projectsPaid tiers lift the five-project cap on Free; shared projects need Team or Enterprise.LimitFree plan: five projectsTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseFree users can create a maximum of five projects.Anthropic: What are projects?20Unlimited projectsPaid tiers lift the five-project cap on Free; shared projects need Team or Enterprise.LimitFree plan: five projectsTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseProject sharing and collaboration Yes Yes YesProject sharing and collaboration No No No NoAnthropic: Plans & pricingFree plan: five projectsPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Show all 4Show fewer
Claude Dashboards18Claude DashboardsClaude writes SQL against a connected warehouse and builds a dashboard from a plain-language question.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude Dashboards is available in beta on Pro, Max, Team, and Enterprise plans.Claude writes a SQL query for each chart and runs it against your data source.Anthropic: Get started with Claude DashboardsPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Video
Claude Motion21Claude MotionShort animations built as code from your content, for explainers and animated charts; no realistic footage.TiersTeam (Standard seat), Team (Premium seat), EnterpriseClaude Motion is available in beta on Team and Enterprise plans.Claude Motion turns your content into short animations, like an explainer for an all-hands, an animated chart for a board deckClaude Motion doesn't use a video generation model, so it doesn't create realistic footage or people.Anthropic: Get started with Claude MotionPro: not includedMax 5x: not includedMax 20x: not includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Images
Claude Design22Claude DesignDesigns, prototypes and one-pagers built from chat with on-canvas editing and the company design system.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude Design is available on Free, Pro, Max, Team, and Enterprise plans, and it's on by default.Claude Design lets you create designs, interactive prototypes, one-pagers, and other visual work by chatting with Claude.with on-canvas editing and your design system includedAnthropic: Get started with Claude DesignPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Voice and audio
Voice mode23Voice modeSpoken conversations on mobile, desktop and web that can use web search and connected tools.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseVoice mode is a beta feature available to all plans (Free, Pro, Max, Team, and Enterprise) on Claude Mobile (iOS and Android), Claude Desktop, and the webUsing web search and the tools you've connectedAnthropic: Use voice modePro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included
Bundled extras
Search past chats24Search past chatsPaid accounts can ask Claude to find and reuse details from earlier conversations.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseSearching past chats is available to users on paid plans (Pro, Max, Team, and Enterprise plans) on the web, Claude Desktop, and Claude Mobile apps.Anthropic: Use Claude’s chat search and memory to build on previous contextPro: includedMax 5x: includedMax 20x: includedTeam (Standard seat): includedTeam (Premium seat): includedEnterprise: included

What each tier adds

  1. Pro$20 a month, or $200 a year billed up frontIncludes
  2. Max 5x$100 a monthEverything in Pro, plus
  3. Max 20x$200 a monthEverything in Pro, plus
  4. Team (Standard seat)$25 a seat a month, $20 billed annuallyIncludes
  5. Team (Premium seat)$125 a seat a month, $100 billed annuallyEverything in Team (Standard seat), plus
  6. Enterprise$20 a seat a month billed annually, plus usage at API ratesEverything in Team (Standard seat), plus

Plan login or API key

What signing in with a Claude plan gives, beside what an Anthropic API key gives.

Coding tools

With the plan login

  • CI on a subscription token27CI on a subscription tokenclaude setup-token mints an OAuth token that runs scripts, CI jobs and the Claude GitHub Action on the plan; it makes model requests only.LimitToken lasts one yearTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseFor CI pipelines, scripts, or other environments where interactive browser login isn't available, generate a one-year OAuth token with claude setup-tokenThis token authenticates with your Claude subscription and requires a Pro, Max, Team, or Enterprise plan. It can only make model requestsAnthropic: Authentication28CI on a subscription tokenclaude setup-token mints an OAuth token that runs scripts, CI jobs and the Claude GitHub Action on the plan; it makes model requests only.LimitToken lasts one yearTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), Enterprisechoose between creating a long-lived token with your Claude subscription and pasting in an API keyAnthropic: Claude Code GitHub Actions
  • Headless runs on the plan29Headless runs on the planSigned in with the plan, claude -p and the Agent SDK draw on plan limits and leave the monthly API credits untouched.TiersMax 5x, Max 20x, Team (Standard seat), Team (Premium seat)When you're signed in with your Claude plan instead, claude -p and Agent SDK usage still draw from your plan's usage limits and don't use your API credits.Anthropic: Monthly API credits for Max and Team plans
Show all 4Show fewer
  • Claude Code on the plan25Claude Code on the planLogging in to Claude Code with the Claude account runs the terminal agent on the plan's included usage, with no API key.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseWith Pro and Max plans, you now have access to both Claude on the web, desktop, and mobile apps and Claude Code in your terminal with one unified subscription.Anthropic: Use Claude Code with your Pro or Max plan26Claude Code on the planLogging in to Claude Code with the Claude account runs the terminal agent on the plan's included usage, with no API key.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseClaude Code is included with every Team plan seat.For Enterprise plans, Claude Code is included with the single Enterprise seat on new and self-serve plans.Anthropic: Use Claude Code with your Team or Enterprise plan
  • IDE extensions included25IDE extensions includedThe same login covers the Claude Code extensions for VS Code, Cursor and JetBrains IDEs, drawing on the plan's limits.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseYour Pro or Max plan also covers Claude Code in supported IDEs, including VS Code, Cursor and other VS Code forks, and JetBrains IDEs like IntelliJ and PyCharm.IDE usage counts toward the same usage limits shared across Claude and Claude Code.Anthropic: Use Claude Code with your Pro or Max plan26IDE extensions includedThe same login covers the Claude Code extensions for VS Code, Cursor and JetBrains IDEs, drawing on the plan's limits.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseYour seat also covers Claude Code in supported IDEsAnthropic: Use Claude Code with your Team or Enterprise plan

With an API key

  • API key wins in Claude Code25API key wins in Claude CodeA set ANTHROPIC_API_KEY moves Claude Code's billing from the plan to the API, and claude -p always uses it.If you have an ANTHROPIC_API_KEY environment variable set on your system, Claude Code will use this API key for authentication instead of your Claude subscriptionAnthropic: Use Claude Code with your Pro or Max plan27API key wins in Claude CodeA set ANTHROPIC_API_KEY moves Claude Code's billing from the plan to the API, and claude -p always uses it.In non-interactive mode ( -p ), the key is always used when present.Anthropic: Authentication

Usage and billing

With the plan login

  • Session and weekly limits20Session and weekly limitsChat, the apps and Claude Code share one allowance that refills every five hours, under a weekly cap that resets at a fixed time.LimitFive-hour session limit plus a weekly limitTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat)Every plan has usage limits that reset on a rolling five-hour session window, and paid plans add weekly limits on top.Your activity across Claude on web, desktop, mobile, and Claude Code all draws from the same pool.Anthropic: Plans & pricing30Session and weekly limitsChat, the apps and Claude Code share one allowance that refills every five hours, under a weekly cap that resets at a fixed time.LimitFive-hour session limit plus a weekly limitTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat)Weekly limits reset at a fixed time each week that is assigned to your account.Anthropic: What is the Pro plan?
  • Max 5x and 20x usage31Max 5x and 20x usageMax sells the Pro session allowance multiplied by five or by twenty.Limit5x or 20x the Pro allowance per sessionTiersMax 5x, Max 20xMax 5x includes five times the Pro plan's per-session usage allowance.Max 20x includes 20 times the Pro plan's per-session usage allowance.Anthropic: What is the Max plan?
Show all 6Show fewer
  • Team seat allowances20Team seat allowancesA Standard seat gets more usage than Pro, and a Premium seat gets five times a Standard seat.LimitPremium seat: 5x a Standard seatTiersTeam (Standard seat), Team (Premium seat)On Team plans, Standard seats give more than Pro and Premium seats give 5x more than Standard.Anthropic: Plans & pricing
  • Usage credits past the limit26Usage credits past the limitWhen a limit runs out, prepaid usage credits keep Claude and Claude Code working at standard API rates up to a spend cap you set.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat)If your organization is on a Team plan or a seat-based Enterprise plan, you can enable usage credits to allow team members to continue working with Claude, Cowork, and Claude Code after reaching their included usage limits.Anthropic: Use Claude Code with your Team or Enterprise plan32Usage credits past the limitWhen a limit runs out, prepaid usage credits keep Claude and Claude Code working at standard API rates up to a spend cap you set.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat)Usage credits allow individuals subscribed to paid Claude plans (Pro, Max 5x, and Max 20x) to continue using Claude after reaching their included usage limits.you can switch to consumption-based pricing at standard API rates and continue your work without interruption.Your usage continues until you reach your monthly spend limitAnthropic: Manage usage credits for paid Claude plans
  • Enterprise bills usage26Enterprise bills usageCurrent Enterprise plans have no per-seat limits; the seat fee buys access and all usage bills at API rates.TiersEnterpriseIf your organization is on a usage-based Enterprise plan (including self-serve Enterprise), there are no per-seat usage limitsAnthropic: Use Claude Code with your Team or Enterprise plan33Enterprise bills usageCurrent Enterprise plans have no per-seat limits; the seat fee buys access and all usage bills at API rates.TiersEnterprisethe seat fee covers access only, and all usage is billed separately at API rates.Anthropic: What is the Enterprise plan?
  • Priority at peak traffic20Priority at peak trafficMax, Team and Enterprise accounts get first call on capacity when traffic peaks.TiersMax 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterprisePriority access at high traffic times No No Yes YesPriority access at high traffic times Yes Yes YesAnthropic: Plans & pricing

With an API key

  • Pay per token20Pay per tokenThe API meters every token: Opus 5.5 costs $4 in and $20 out per million, Fable 5.1 costs $10 and $50.LimitOpus 5.5: $4 input, $20 output per MTokOpus 5.5 Daily driver for agentic coding and enterprise work Prompt caching Read $0.20 / MTok Write $5 / MTok Input $4 / MTok Output $20 / MTokFable 5.1 Next generation intelligence for long-running agents Prompt caching Read $0.25 / MTok Write $12.50 / MTok Input $10 / MTok Output $50 / MTokAnthropic: Plans & pricing
  • Batch API at half price35Batch API at half priceThe Message Batches API runs large jobs asynchronously at half the token price, most finishing within an hour.Limit50% off input and output tokenswith most batches finishing in less than 1 hour while reducing costs by 50% and increasing throughput.Anthropic: Batch processing36Batch API at half priceThe Message Batches API runs large jobs asynchronously at half the token price, most finishing within an hour.Limit50% off input and output tokensThe Batch API allows asynchronous processing of large volumes of requests with a 50 percent discount on both input and output tokens.Anthropic: Pricing
Show all 5Show fewer
  • API billed apart30API billed apartAPI keys come from the Claude Console and bill on their own; the Pro plan includes no API usage at all.The Pro plan does not include API usage through the Claude Console.Anthropic: What is the Pro plan?34API billed apartAPI keys come from the Claude Console and bill on their own; the Pro plan includes no API usage at all.A paid Claude subscription enhances your chat experience, and the Claude API is billed separately.The Claude Console is our developer platform providing API keys and access to Claude models for building applications and integrations.Anthropic: I have a paid Claude subscription (Pro, Max, Team, or Enterprise plans). Why is Claude API usage billed separately from my paid Claude plan?
  • Priority Tier closed37Priority Tier closedAnthropic no longer sells Priority Tier commitments; existing contracts keep it, and guaranteed capacity goes through sales.Priority Tier capacity commitments are no longer available for purchase. Organizations with an existing commitment can continue to use Priority Tier through their contract end dateIf you need guaranteed capacity, contact salesAnthropic: Service tiers
  • Spend tiers set limits38Spend tiers set limitsAPI organizations move up usage tiers automatically with history, each with a monthly spend cap.LimitStart $500, Build $1,000, Scale $200,000 a monthOrganizations are placed on a tier automatically based on usage history and account standingUsage tier Monthly spend cap Start $500 USD Build $1,000 USD Scale $200,000 USDAnthropic: Rate limits

Models

With the plan login

  • Fable inside the limits39Fable inside the limitsMax plans and Team Premium seats run Fable 5 and Fable 5.1 at no extra cost on up to half the weekly limit.LimitUp to 50% of the weekly limit on FableTiersMax 5x, Max 20x, Team (Premium seat)On Max plans, premium seats on Team plans, and premium seats on legacy seat-based Enterprise plans, both are a standard part of your plan.You can use up to 50% of your weekly usage limits on Fable models at no extra cost.Anthropic: Claude Fable models on your plan
  • Fable on credits for Pro39Fable on credits for ProPro plans and Team Standard seats reach Fable models only through pay-as-you-go usage credits.TiersPro, Team (Standard seat)Pro plans and standard seats on Team plans: Fable 5 and Fable 5.1 aren't included in your plan's usage limits. You can use them with usage creditsAnthropic: Claude Fable models on your plan
Show all 4Show fewer
  • Opus from Pro up20Opus from Pro upEvery paid tier reaches Opus; the Free plan stops at Sonnet and Haiku.TiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseOpus No Yes Yes Yes Sonnet Yes Yes Yes YesOpus Yes Yes Yes Sonnet Yes Yes YesAnthropic: Plans & pricing
  • 1M-token context401M-token contextOn paid plans the newest models take up to a million tokens of context in one chat.LimitUp to 1M tokens, 500K or 200K on older modelsTiersPro, Max 5x, Max 20x, Team (Standard seat), Team (Premium seat), EnterpriseOn paid plans, the newest models support up to a 1M token context window, while others support 500K or 200K tokens.Anthropic: How do usage and length limits work?

With an API key

  • Fast mode for Opus20Fast mode for OpusFast mode serves Opus 5.5 up to 2.5 times faster at double the standard price, outside the Batch API.Limit2x standard pricingGet up to 2.5x faster speeds with fast mode for Opus 5.5 at 2x standard pricing.Anthropic: Plans & pricing36Fast mode for OpusFast mode serves Opus 5.5 up to 2.5 times faster at double the standard price, outside the Batch API.Limit2x standard pricingFast mode is not available with the Batch APIAnthropic: Pricing
  • Models on cloud platforms41Models on cloud platformsFable 5.1 and the 5.5 models also run on Amazon Bedrock, Google Cloud and Microsoft Foundry under their own model IDs.Amazon Bedrock ID anthropic.claude-fable-5-1 anthropic.claude-opus-5-5 anthropic.claude-sonnet-5-5 anthropic.claude-haiku-5-5Microsoft Foundry ID claude-fable-5-1 claude-opus-5-5 claude-sonnet-5-5 claude-haiku-5-5Claude Models on Google Cloud (Google managed) ID claude-fable-5-1Anthropic: Models overview

Building on it

With the plan login

  • Monthly API credits29Monthly API creditsMax and Team plans add a monthly Claude Platform credit for the API, the Agent SDK and claude -p run with an API key from the linked Console organization; it expires each cycle and never pays for interactive Claude Code.Limit$100 on Max 5x, $200 on Max 20x, $20 or $100 a Team seat up to $500 pooledTiersMax 5x, Max 20x, Team (Standard seat), Team (Premium seat)Max 5x $100 Max 20x $200 Team (Standard seats) $20 per seat* Team (Premium seats) $100 per seat*On Team plans, credits for all seats are pooled into one monthly balance, capped at $500.Doesn't roll over. Unused credits expire at the end of each billing cycle.Use them to build Claude into your own apps and agents with the Claude API.The credits don't apply to: Interactive Claude Code in the terminal, IDE, desktop, or webAPI credits cover claude -p and the Claude Agent SDK when you run them yourself with an API key from your linked Claude Console organization, because that usage is billed as Agent SDK usage.Anthropic: Monthly API credits for Max and Team plans

With an API key

  • Prompt caching42Prompt cachingCached prompt prefixes read back at 0.05x the input price on Opus 5.5 and Sonnet 5.5, 0.025x on Fable 5.1 and Mythos 5.1, and 0.1x on other models; writes cost 1.25x for five minutes or 2x for an hour.LimitReads at 0.05x on Opus 5.5 and Sonnet 5.5, 0.025x on Fable 5.1 and Mythos 5.1, 0.1x on other modelsCache write tokens with a 5-minute TTL are 1.25 times the base input tokens price. Cache write tokens with a 1-hour TTL are 2 times the base input tokens price. Cache read tokens are 0.1 times the base input tokens price (see the table footnote for per-model exceptions).Cache hits and refreshes on Claude Opus 5.5 and Claude Sonnet 5.5 are priced at 0.05x the base input price.Cache hits and refreshes on Claude Fable 5.1 and Claude Mythos 5.1 are priced at 0.025x the base input price.Anthropic: Prompt caching
  • Managed Agents20Managed AgentsHosted agent sessions bill standard token rates plus a runtime fee for each active session-hour.Limit$0.08 per active session-hourBuild and deploy agents at scale with a suite of composable APIs. Standard token rates apply. Cost $0.08 per session-hour for active runtimeAnthropic: Plans & pricing
Show all 3Show fewer
  • Web search and code tools20Web search and code toolsAPI calls can use server-side web search at $10 per thousand searches and a sandboxed code execution tool.Limit50 free container hours a day, then $0.05 an hourGive Claude access to the latest information from the web. Doesn't include input and output tokens required to process requests. Cost $10 / 1K searchesRun Python code in a sandboxed environment for advanced data analysis.50 free hours of usage daily per organization. Additional hours $0.05 per hour per containerAnthropic: Plans & pricing

Data and admin

With the plan login

  • No training on business plans20No training on business plansTeam and Enterprise content stays out of model training by default; Pro and Max users must opt out themselves.TiersTeam (Standard seat), Team (Premium seat), EnterpriseModel training More information None by default None by default None by defaultModel training More information Opt-out Opt-out Opt-out Opt-outAnthropic: Plans & pricing
  • Enterprise compliance controls33Enterprise compliance controlsEnterprise adds SCIM, audit logs and custom retention, plus a Compliance API that exports chats and activity logs.TiersEnterpriseAudit logs: capture key information about user actions, system events, and data access. System for Cross-domain Identity Management (SCIM) Custom data retention controlsCompliance API: programmatically access Claude usage data including activity logs, chat histories, and file contentAnthropic: What is the Enterprise plan?

With an API key

  • Zero data retention43Zero data retentionUnder a ZDR arrangement, set up through sales for each organization, Anthropic stores no prompts or responses after the reply.Under a ZDR arrangement, Anthropic does not store customer prompts or responses at rest after the API response is returned. To request ZDR for your organization, contact the Anthropic sales teamZDR is enabled per organizationAnthropic: API and data retention
  • US-only inference20US-only inferenceRequests can be pinned to US data centers for a 10% surcharge on input and output tokens.Limit1.1x token pricingFor workloads that need to run in the US, US-only inference is available at 1.1x pricing for input and output tokens.Anthropic: Plans & pricing

More on Claude

  1. GILD ReviewGILD's evaluation, with the buying facts
  2. Research3 pieces
  3. Guides4 guides
  4. Operator Insights10 operator accounts
  5. Where It FitsOn the AI Knowledge Work stack
  6. Related Products7 products