Free $0
“Free covers everyday questions.” 5-hour rolling session limits.
Projects: “Up to 5.” “Claude Design, Slides, Docs”. No Claude Code or Research. “Fable 5 isn’t available on the Free plan.”
Live tracker Last checked 2026-10-16 Primary sources only
Every AI plan's real price and limits, on one page.
Email me if it changes
One email per change, plus my Tuesday note. Unsubscribe anytime.
You're on the list. I'll email you when plans change.
Vendors change plans quietly and their pages don't keep history. This one does.
Claude, ChatGPT, Gemini, Microsoft Copilot, Grok and Perplexity. Every number below was read off the vendor's own page on the date shown, their limit wording quoted as written, and every change logged with a date. If a number has aged, the source page wins. Tell me and I'll log the change.
Writing about this? Link the page. The dated changelog is the part your readers can't get from the vendor.
Which plan should you buy?
Three questions. Your verdict is free.
● Your plan
ChatGPT Plus · $20/mo
One $20 subscription for general work: ChatGPT Plus. It's more general, you'll use it for more things, and its limits bend instead of break.
Before you subscribe
ChatGPT Plus will still forget you tomorrow. Every chat starts from zero.
The Setup is one folder. Answer three questions and it writes your memory file and your voice file, so every chat starts already knowing you. Twenty minutes, no code.
See The Setup → $39 · 30-day refund 01
Claude (Anthropic)
Pro $20/mo ($17/mo billed annually at $200 up front)
“Pro gives you at least 5x more usage per 5-hour session than Free.”
Claude Code, Claude Design, Slides, Docs, Science, Projects, Research. “Claude Dashboards (beta)”. “Hand off and schedule tasks.” Pricing table: “Up to 1M” context, “varies by model.” “Fable 5 and Fable 5.1 aren’t included in your plan's usage limits. You can use them with usage credits.”
Max 5x $100/mo
“Max 5x includes five times the Pro plan's per-session usage allowance.” Plus a weekly limit across all models.
“Fable 5 and Fable 5.1 are included as a standard part of your plan. You can use up to 50% of your weekly usage limits on Fable models at no extra cost.” Higher output limits, priority at peak. “Monthly API credits”: “$100/month”. Claim after seven days on an eligible plan by linking one Console organization; credits expire each billing cycle and do not increase Claude or Claude Code plan limits.
Max 20x $200/mo
“Max 20x includes 20 times the Pro plan's per-session usage allowance.” Plus a weekly limit across all models.
“Fable 5 and Fable 5.1 are included as a standard part of your plan. You can use up to 50% of your weekly usage limits on Fable models at no extra cost.” Early access to advanced features. “Monthly API credits”: “$200/month”. Claim after seven days on an eligible plan by linking one Console organization; credits expire each billing cycle and do not increase Claude or Claude Code plan limits.
Team $25/seat/mo standard ($20 annual) · $125/seat/mo premium ($100 annual)
Standard seats: more usage than Pro. Premium seats: “5x more usage than standard seats.”
“Claude Motion and Dashboards (beta)”. Pricing table: “Up to 1M” context, “varies by model.” Standard seats use Fable 5 and Fable 5.1 with usage credits; on premium seats, both are included as a standard part of the plan, up to 50% of weekly usage limits at no extra cost. “Up to $500 per month in pooled API credits”. Standard seats contribute $20 each and Premium seats $100 each; claim after seven days by linking one Console organization.
Enterprise $20/seat/mo, billed annually, 20 seat minimum, plus usage at API rates
“Seat price + usage at API rates.” “Usage cost scales with model and task.”
“Claude Security (beta)”. “All Team plan features, plus:” user and organization spend limits, role-based access, SCIM, audit logs, Compliance API, custom data retention, network-level access control, IP allowlisting, and an available HIPAA-ready offering.
Education plan Discounted rates (no numeric price published)
“Comprehensive access for all university members at discounted rates.”
“Get a comprehensive university‐wide plan for an institution, including its students, faculty, and staff.” Dedicated API credits, educational features, training and enablement.
Notes
Anthropic's own line: “Every plan has usage limits that reset on a rolling five-hour session window, and paid plans add weekly limits on top.” Pro and Max reset every five hours and also carry a weekly cap. Enterprise usage is billed at API rates. Context: the pricing table says “Up to 1M” and “varies by model.” On paid plans, Fable 5.1, Opus 5.5, Opus 5, Sonnet 5.5, Sonnet 5 and Haiku 5.5 chat at 1M; Fable 5, Opus 4.8 to 4.6 and Sonnet 4.6 at 500K. Claude Code and Cowork have their own 1M exceptions; Haiku 5.5 has 1M in Claude Code and 500K in Cowork.
Deciding, not just comparing? Claude Pro vs Max, from someone who pays for the 20x tier
02
ChatGPT (OpenAI)
Free $0
“Unlimited text chats with GPT-5.6 Luna,” subject to abuse guardrails.
No GPT-5.6 Sol. Limited uploads, image gen, voice, deep research, Codex. 27K context.
Go $8/mo (corrected Sep 5, 2026, read off OpenAI's rendered pricing page; an earlier $12 entry here was wrong)
“Free and Go users both have unlimited everyday text chats,” plus higher limits than Free for image gen, uploads, data analysis.
No GPT-6 Astra, no GPT-5.6 Sol, no legacy models. Reasoning via Think on GPT-5.6 Luna. Ad-eligible. 54K/256K context, identical to Plus.
Plus $20/mo
Standard Plus pools (OpenAI publishes no fixed counts).
GPT-5.6 Sol yes, Sol Pro no. Deep research, scheduled tasks, record mode, expanded Codex. 54K/256K context.
Pro (three tiers) Pro 100: $100/mo · Pro 200: $200/mo · Pro 500: $500/mo. “Pro 200 is also available for new subscriptions again.”
“Pro $100 unlocks 5x higher usage than Plus, while Pro $200 unlocks 20x usage than Plus.” “New subscriptions that aren’t eligible for grandfathering include a lower usage allowance.” “Eligible customers can use the previous included allowance through October 29, 2026 while their Pro 200 subscription is active.”
GPT-5.6 Sol Pro included. Sol/Terra/Luna “unlimited*”. 128K/400K context. Ultrafast: “Not included” on Pro 100 and Pro 200; “Included” on Pro 500. “GPT-6 Astra Ultrafast and GPT-6.1 Sol Ultrafast.”
Notes
OpenAI's pricing pages render prices client-side; the numbers below come from OpenAI's own help-center plan articles. No annual billing exists for Go, Plus, or Pro.
Deciding, not just comparing? Claude Max vs ChatGPT Pro, I pay for both
03
Gemini (Google)
Free $0
5-hour refresh windows up to a weekly limit.
“Access to our Flash-Lite model”. Image generation and editing, Deep Research, Gemini Live, Canvas and Gems.
Google AI Plus $4.99/mo
“2x higher usage limits than Free.”
“Access to our Flash model”. “Get 200 Google Flow Credits” with access to Gemini Omni Flash and custom tool creation. Gemini Notebook, Gemini in Gmail/Vids, video generation and Daily Brief, 400 GB storage.
Google AI Pro $19.99/mo
“4x higher usage limits than Free.”
“Access to Gemini 4 Argon”. “Access to our Pro model”. “Access to our Flash model”. Deep Research, Gemini in Gmail/Docs/Vids, Flow (1,000 Google Flow Credits), 5 TB, YouTube Premium Lite.
Google AI Ultra 5x $99.99/mo
“5x higher usage limits vs. AI Pro.”
“Access to Gemini 4 Argon”. “Highest access to our Pro model”. “Highest access to our Flash model”. “first access to advanced features like Tasks in Gemini (in select countries).” Project Genie, “10,000 or 25,000 Google Flow Credits,” storage starting at 20 TB, YouTube Premium.
Google AI Ultra 20x $199.99/mo
“20x higher usage limits vs. AI Pro.”
“Access to Gemini 4 Argon”. “Highest access to our Pro model”. “Highest access to our Flash model”. “first access to advanced features like Tasks in Gemini (in select countries).” Project Genie, “10,000 or 25,000 Google Flow Credits,” storage starting at 20 TB, YouTube Premium.
Notes
Google's official limit language: limits “factor in the complexity of your prompt, the features you use, and the length of your chat,” refreshing every 5 hours until a weekly limit.
Deciding, not just comparing? Gemini vs ChatGPT: your job probably already bought you one
04
Microsoft Copilot
Copilot (free) Free
“Subject to available capacity and limits.”
“Microsoft offers a limited free version of Copilot subject to available capacity and limits.” “For deeper integration across Microsoft 365 apps like Word, Excel, PowerPoint, and Outlook, Copilot is included with eligible Microsoft 365 subscriptions.”
Microsoft 365 Personal $9.99/mo or $99.99/yr
Overall usage: “Higher than free.” AI image generation: “Higher than free.”
Copilot in Word/Excel/PowerPoint/Outlook desktop. 1 TB.
Microsoft 365 Family $12.99/mo or $129.99/yr
Overall usage: “Higher than free.” AI image generation: “Higher than free.”
“Everything in Microsoft 365 Personal, plus: For 1 to 6 people (AI for subscription owner only).” “2 TB of shared, secure cloud storage”.
Microsoft 365 Premium $19.99/mo or $199.99/yr
Overall usage: “Extensive.” AI image generation: “Extensive.”
Advanced Copilot chat with extensive usage; limited usage for agentic AI that gets things done for you including advanced research and analytics; preferred access to AI models during peak times.
Microsoft 365 Pro $99.99/mo (monthly billing only)
Overall usage: “Highest.” AI image generation: “Highest.”
“Everything in Microsoft 365 Premium, plus:” Advanced Copilot chat with highest usage; highest usage for agentic AI that gets things done for you including advanced research and analytics; highest usage for visual AI assistance from your screen or camera.
Copilot Chat (work) Included with an eligible Microsoft 365 subscription
“Access and use agents, priced on a metered basis.”
“Get secure, enterprise-ready AI chat, available at no additional cost for all Microsoft Entra account users with an eligible Microsoft 365 subscription.” An Azure subscription is required to use agents; prepaid Copilot Studio capacity packs are also available.
Microsoft 365 Copilot (work seat) $30.00/user/mo, paid yearly
“Annual subscription–auto renews.” A separate qualifying Microsoft 365 plan license is required.
“Everything in Copilot Chat, plus:” AI chat grounded in business data; Copilot in Word, Excel, PowerPoint, Outlook, and Microsoft Teams; agents with Copilot Studio.
Notes
Two audiences. The individual page and its linked plan comparison cover Copilot Free plus Microsoft 365 Personal, Family, Premium and Pro. The enterprise page lists Copilot Chat as included and Microsoft 365 Copilot at $30 per user per month, paid yearly. Consumer AI features are owner-only, and “usage limits apply.”
Deciding, not just comparing? Copilot vs ChatGPT: first find out which Copilot you have
05
Grok (xAI)
Free $0/mo
“Limited” AI usage. “Limited Chat requests”. “Limited agent requests”.
Chat, image generation and Grok Build: “Limited”. Video generation, Grok API access and Grok Bot: “Not included”.
Plus $8/mo
“Basic” usage. “Shared” AI usage pool.
“Grok Chat, Voice, Imagine, and Build”. “Grok API access”. “Cursor”. Video generation and Grok Bot: “Not included”.
Premium $30/mo
“4x” “more usage compared to Plus”.
“Everything in Plus, and:” Grok Bot, latest Grok models, faster Chat responses, image and video generation, code hosting and AI code reviews, cloud agents and automations.
Super $100/mo
“15x” “more usage compared to Plus”.
“Everything in Premium, and:” “Priority model access at peak times”.
Ultra $200/mo
“50x Plus” usage.
Grok Chat, image and video generation, Grok API access, Grok Build and Grok Bot: “Included”. “Priority model access at peak times” and “Earliest access to all products & models”: “Included”.
Teams $30+/user/mo
“4-50x Plus” usage. “Individual” “usage pool shared across products”.
“Team marketplace for internal rules, skills, and plugins”. “Centralized billing and administration”. “Usage analytics”. “Team-wide privacy settings”. “Custom Chat data retention”. “SAML/OIDC SSO”. “Mix and match seat types”.
Enterprise Custom (contact sales)
“Pooled” usage. “Shared” “usage pool across all team members”.
“Everything in Teams, and:” “Invoice/PO billing”, “SCIM seat management”, “Granular spend controls”, group-level spend limits, auto-run/browser/network controls, repository/model/MCP access controls, audit logs and service accounts, and access/network/audit controls for Grok Bot.
Notes
x.ai/pricing now lists Free, Plus, Premium, Super, Ultra, Teams and Enterprise. “Every tier includes a pool of AI usage shared across Grok and Cursor. Within Grok, the pool covers Chat, Imagine, Voice, and Bot.” “Your allowance refreshes weekly”. Top-ups are available. “API usage is billed separately at standard model rates and doesn’t draw from your plan’s usage pool.” Third-party models in Cursor are billed separately.
Deciding, not just comparing? What is Grok Bot? A paying user's week-one review
06
Perplexity
Free $0
“Good for limited daily usage.” Web answers, uploads, asset gen, Comet Agent all “Limited.”
Basic AI models. Citations in every answer.
Pro $20/mo
10x web answers, uploads, asset generation, and Comet Agent vs Free.
Deep Research, top models + model selection, 4,000 bonus credits, 400+ connectors.
Max $200/mo
“Highest usage limits and memory,” priority access.
Frontier models, Model Council, Brain, 35,000 bonus credits + 10,000 monthly credits, 5x video gen.
Notes
Monthly pricing only shown on the official page.
Deciding, not just comparing? Perplexity vs ChatGPT: keep it free, pay the other one
ALERTS
When a plan moves, you get one email.
19 plan changes since August. Nobody at Anthropic, OpenAI or xAI emailed anyone about any of them. I read the pages every morning.
Make your limit last
I burned 50% of my weekly Claude limit in one day. On the biggest plan.
Get the 4 fixes I use now so it lasts the week. Free, shows up right here.
LOG
Every plan change, dated
The part vendor pages don't give you: what changed and when. Entries before 2026-08-21 come from dated observations and screenshots in our published reviews.
- 2026-10-16
Grok: replaced the old SuperGrok ladder with the current public Free $0, Plus $8, Premium $30, Super $100, Ultra $200, Teams $30+/user and custom Enterprise tiers; corrected Bot to the shared weekly Grok/Cursor pool and kept API usage separate (https://x.ai/pricing). Google: Free now lists Flash-Lite, Plus lists Flash, Pro and Ultra include Gemini 4 Argon, and Ultra lists Tasks in Gemini instead of the old Deep Think/Spark wording (https://gemini.google/subscriptions/). Anthropic: added Free Design/Slides/Docs, Max $100/$200 monthly API credits, Team pooled credits up to $500, and Enterprise’s 20-seat minimum (https://claude.com/pricing; https://support.claude.com/en/articles/17154008-monthly-api-credits-for-max-and-team-plans). Claude, Gemini, Microsoft Copilot and Grok rows were verified today. ChatGPT pricing and Perplexity were unreadable; their dates remain unchanged. Verdict-sensitive picker posts remain held for review.
- 2026-10-09
OpenAI: corrected Pro to three monthly tiers, Pro 100 $100, Pro 200 $200 and Pro 500 $500; Pro 200 is open to new subscriptions again, eligible customers keep their previous allowance through October 29, and only Pro 500 includes Ultrafast (https://help.openai.com/en/articles/9793128-what-is-chatgpt-pro). Microsoft: corrected Family storage to 2 TB shared (https://www.microsoft.com/en-us/microsoft-365/buy/compare-all-microsoft-365-products). Anthropic: added Haiku 5.5 to paid-chat 1M context, kept its Cowork context at 500K, added Dashboards, Motion and Security beta gates, and updated exact matching model lists in two picker posts (https://support.claude.com/en/articles/8606394-how-large-is-the-context-window-on-paid-claude-plans; https://claude.com/pricing). Claude, Gemini and Microsoft Copilot rows now match the morning audit. ChatGPT remains partly unverified; Grok and Perplexity were not fully verified.
- 2026-10-02
Anthropic: added Sonnet 5.5 to the paid-plan 1M chat context list and updated exact matching model lists in two picker posts; the support table lists “Claude Sonnet 5.5” with “1M tokens” (https://support.claude.com/en/articles/8606394-how-large-is-the-context-window-on-paid-claude-plans). Claude, Gemini and Microsoft Copilot prices and listed limits were rechecked; no dollar prices changed. ChatGPT, Grok and Perplexity were not fully verified this run.
- 2026-09-25
Anthropic: corrected Free Projects to up to five and the pricing-table context wording to up to 1M, varying by model; added Opus 5.5 at 1M and Fable 5 at 500K for paid chat, current Pro task/tool wording, and the unpriced Education plan (https://claude.com/pricing; https://support.claude.com/en/articles/8606394-how-large-is-the-context-window-on-paid-claude-plans). Google: replaced the unsupported AI Plus Chrome claim with its listed video generation and Daily Brief access (https://gemini.google/subscriptions/). Microsoft: added the included Copilot Chat work tier and aligned Family/work-seat wording (https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/individuals; https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/enterprise). No dollar prices changed.
- 2026-09-22
OpenAI: GPT-6 Sol and GPT-6 Luna are out. OpenAI's line: they “roll out today in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu users.” Not in Chat, the same split Astra had on September 3. Free and Go get Luna in the desktop app, per launch coverage (the help center still describes GPT-5.6). No ChatGPT usage numbers published for either model. Sources: https://openai.com/index/introducing-gpt-6-sol-and-luna/ and https://community.openai.com/t/announcing-gpt-6-sol-and-gpt-6-luna/1399925.
- 2026-09-22
Anthropic: Claude Opus 5.5 is out and listed on Pro, Max 5x, Max 20x and Team (not Free). Anthropic's own line: it “performs at the level of Claude Fable 5.1 on most work.” Same day: “we're increasing five-hour usage limits on Pro, Max, Team, and seat-based Enterprise plans,” plus a one-time rate limit reset subscribers can “save and use whenever you choose.” No new limit numbers published. Opus 5 stays available as a legacy model. Sources: https://www.anthropic.com/claude-opus-5-5 and https://claude.com/pricing.
- 2026-09-11
xAI: the full SuperGrok ladder is now on the tracker. Lite $10/mo, Heavy $300/mo and Business $30/seat added alongside Free, SuperGrok $30 and SuperGrok Plus $100. These three are NOT on x.ai/pricing, which lists Lite and Heavy as comparison-table columns with no number; they print on the signed-in subscription screen at grok.com, read off my own paying account and screenshotted in the post. An earlier note here said no published price existed for “Lite or Heavy”, true of the public page, and it also got tangled with the separately-unpriced Grok Bot agent tiers. Both now stated precisely.
Show the other 15 changes
- 2026-09-11
OpenAI: new sign-ups and upgrades to ChatGPT Pro $200 are paused as of September 10, 2026, after demand for GPT-6 Astra outran capacity. Existing $200 subscriptions keep renewing and Pro $100, Plus, Go and API are unaffected (help.openai.com).
- 2026-09-11
Microsoft: added Microsoft 365 Pro at $99.99/month (monthly billing only); corrected Personal and Family AI image generation to “Higher than free” and Premium overall/image usage from “Highest” to “Extensive.” The enterprise work seat remains $30/user/month, paid yearly. Sources: https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/individuals and https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/enterprise.
- 2026-09-11
Anthropic: expanded the context note with the current paid-chat 500K exceptions for Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6, plus the separate Claude Code and Cowork 1M exceptions. Source: https://support.claude.com/en/articles/8606394-how-large-is-the-context-window-on-paid-claude-plans.
- 2026-09-05
CORRECTION: our ChatGPT Go row said $12/mo. Go is $8/mo, read off OpenAI's own rendered pricing page (it serves plan prices client-side, so scrapes miss them). Also corrected: Go does have reasoning, via Think on GPT-5.6 Luna; it lacks GPT-6 Astra and GPT-5.6 Sol. Go and Plus share the same 54K/256K context window.
- 2026-09-04
Anthropic: added the Enterprise plan at $20 per seat per month, billed annually, plus usage at API rates; the page says “Usage cost scales with model and task.” Source: https://claude.com/pricing.
- 2026-09-01
Microsoft: added Microsoft 365 Family at $12.99/month or $129.99/year, with AI available to the subscription owner only, from https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/individuals.
- 2026-09-01
Google: split AI Ultra's $99.99 5x and $199.99 20x options, corrected its Flow-credit wording to “10,000 or 25,000,” added AI Plus's 200 Flow credits, and removed Deep Think and Gemini Spark from AI Pro based on https://gemini.google/subscriptions/.
- 2026-09-01
Anthropic: Fable 5.1 now follows Fable 5's plan inclusion rules, and Fable 5.1, Opus 5, and Sonnet 5 support a 1M paid-plan chat context window; sources: https://support.claude.com/en/articles/15424964-claude-fable-5-on-your-plan and https://support.claude.com/en/articles/8606394-how-large-is-the-context-window-on-paid-claude-plans.
- 2026-08-29
xAI: Grok Bot (launched August 11, early beta) added to the tracker. “Grok Bot access” is listed on the $30 SuperGrok tier per x.ai/pricing (checked August 28), with its own usage pool separate from chat plans per the launch post. No published price for Bot Lite or Heavy. Launch-week Reddit claims of $120/seat or a $200–300 requirement are stale, full breakdown in our week-one review.
- 2026-08-21
Anthropic's 50% weekly-usage boost on paid plans (running since May 13) reported extended again through August 31, its third extension, with Anthropic saying it hopes to make the allowance permanent.
- 2026-08-21
Baseline snapshot published. Every number re-read off the official pages today. Notable find: Microsoft's standalone Copilot Pro consumer plan is gone, the old pricing URL 404s and consumer Copilot now ships inside Microsoft 365 Personal/Family/Premium.
- 2026-08-07
Anthropic: Claude Cowork listed as beta on web and mobile for Max, Team, and Enterprise, “rolling out to Pro plans over the next several weeks” (Anthropic's own wording at the time).
- 2026-07-31
Google's consumer lineup observed as Free / AI Plus / AI Pro / AI Ultra (screenshotted for our Gemini review). Claude pricing page observed at Free $0 / Pro $17–20 / Max from $100.
- 2026-07-27
ChatGPT Go observed at $12/mo with the official note “this plan may include ads” (documented in our Go vs Plus review). See the 2026-09-05 entry: the $12 figure was wrong.
- 2026-07-20
Anthropic: Fable 5 moved OUT of Claude Pro's plan usage, it now bills as separate usage credits on Pro, and is included on Max capped at 50% of weekly limits. The single biggest plan-value change of the summer.
The dataset
The full price history ships as a CC-BY CSV on GitHub: ai-plan-pricing-history . Use it, cite it, build on it.