Skip to content
Best AI Ever

Subscriptions

AI subscriptions compared

What you get for your money in the ChatGPT, Claude, Gemini, Grok, Mistral Vibe, Perplexity and Copilot apps: prices in US dollars and euros as the providers publish them, the models included, the published usage limits and the main features. Plans are scored against others in the same price range.

Plans and prices verified on 2026-11-01 · 31 plans from 7 providers

Quick ranking

Pick a price and a use, and the ranking shows here. Plans are scored only against others in the same price range; the full tables, plan details and how we score follow below.

#PlanScorePrice (US)Price (DE)
#1 Gemini FreeGoogle 76.0 $0.00 0.00 €
#2 Claude FreeAnthropic 71.1 $0.00 –
#3 ChatGPT FreeOpenAI 71.1 $0.00 0.00 €
#4 Mistral Vibe FreeMistral AI 67.0 $0.00 –
#5 Copilot FreeMicrosoft 66.7 $0.00 0.00 €
#6 The X Subscription FreexAI 66.2 $0.00 –
#7 Perplexity FreePerplexity 65.0 $0.00 –

All criteria for free plans

#PlanScorePrice (US)Price (DE)
#1 Gemini FreeGoogle 78.2 $0.00 0.00 €
#2 Claude FreeAnthropic 74.9 $0.00 –
#3 ChatGPT FreeOpenAI 72.6 $0.00 0.00 €
#4 Copilot FreeMicrosoft 72.2 $0.00 0.00 €
#5 The X Subscription FreexAI 71.0 $0.00 –
#6 Mistral Vibe FreeMistral AI 68.8 $0.00 –
#7 Perplexity FreePerplexity 66.7 $0.00 –

All criteria for free plans · Best AI subscription for work

#PlanScorePrice (US)Price (DE)
#1 Gemini FreeGoogle 76.7 $0.00 0.00 €
#2 ChatGPT FreeOpenAI 72.6 $0.00 0.00 €
#3 Claude FreeAnthropic 72.1 $0.00 –
#4 Mistral Vibe FreeMistral AI 68.8 $0.00 –
#5 The X Subscription FreexAI 68.3 $0.00 –
#6 Copilot FreeMicrosoft 66.7 $0.00 0.00 €
#7 Perplexity FreePerplexity 61.9 $0.00 –

All criteria for free plans · Best AI subscription for coding

#PlanScorePrice (US)Price (DE)
#1 Gemini FreeGoogle 76.7 $0.00 0.00 €
#2 ChatGPT FreeOpenAI 72.6 $0.00 0.00 €
#3 Claude FreeAnthropic 72.1 $0.00 –
#4 The X Subscription FreexAI 70.0 $0.00 –
#5 Mistral Vibe FreeMistral AI 68.8 $0.00 –
#6 Copilot FreeMicrosoft 66.7 $0.00 0.00 €
#7 Perplexity FreePerplexity 66.7 $0.00 –

All criteria for free plans · Best AI subscription for research

#PlanScorePrice (US)Price (DE)
#1 Google AI ProGoogle 94.0 $19.99 21.99 €
#2 ChatGPT PlusOpenAI 86.6 $20.00 23.00 €
#3 Claude ProAnthropic 86.5 $20.00 –
#4 Perplexity ProPerplexity 83.9 $20.00 –
#5 Microsoft 365 PremiumMicrosoft 76.7 $19.99 22.00 €
#6 The X Subscription PremiumxAI 74.3 $30.00 31.00 €

All criteria for standard plans (about 20 a month)

#PlanScorePrice (US)Price (DE)
#1 Google AI ProGoogle 94.0 $19.99 21.99 €
#2 Claude ProAnthropic 90.5 $20.00 –
#3 ChatGPT PlusOpenAI 88.6 $20.00 23.00 €
#4 Perplexity ProPerplexity 86.9 $20.00 –
#5 Microsoft 365 PremiumMicrosoft 81.9 $19.99 22.00 €
#6 The X Subscription PremiumxAI 74.3 $30.00 31.00 €

All criteria for standard plans (about 20 a month) · Best AI subscription for work

#PlanScorePrice (US)Price (DE)
#1 Google AI ProGoogle 94.0 $19.99 21.99 €
#2 Claude ProAnthropic 90.5 $20.00 –
#3 ChatGPT PlusOpenAI 88.6 $20.00 23.00 €
#4 Perplexity ProPerplexity 81.2 $20.00 –
#5 Microsoft 365 PremiumMicrosoft 74.8 $19.99 22.00 €
#6 The X Subscription PremiumxAI 74.3 $30.00 31.00 €

All criteria for standard plans (about 20 a month) · Best AI subscription for coding

#PlanScorePrice (US)Price (DE)
#1 Google AI ProGoogle 94.0 $19.99 21.99 €
#2 Claude ProAnthropic 90.5 $20.00 –
#3 ChatGPT PlusOpenAI 88.6 $20.00 23.00 €
#4 Perplexity ProPerplexity 86.9 $20.00 –
#5 Microsoft 365 PremiumMicrosoft 79.5 $19.99 22.00 €
#6 The X Subscription PremiumxAI 74.3 $30.00 31.00 €

All criteria for standard plans (about 20 a month) · Best AI subscription for research

#PlanScorePrice (US)Price (DE)
#1 Google AI Ultra (5x)Google 94.0 $99.99 99.99 €
#2 Claude MaxAnthropic 90.5 $100.00 –
#3 ChatGPT Pro 100OpenAI 88.6 $100.00 103.00 €
#4 Microsoft 365 ProMicrosoft 83.3 $99.99 110.00 €
#5 The X Subscription SuperxAI 81.0 $100.00 105.00 €

All criteria for advanced plans (about 100 a month)

#PlanScorePrice (US)Price (DE)
#1 Claude MaxAnthropic 94.5 $100.00 –
#2 Google AI Ultra (5x)Google 94.0 $99.99 99.99 €
#3 ChatGPT Pro 100OpenAI 90.6 $100.00 103.00 €
#4 Microsoft 365 ProMicrosoft 90.0 $99.99 110.00 €
#5 The X Subscription SuperxAI 81.0 $100.00 105.00 €

All criteria for advanced plans (about 100 a month) · Best AI subscription for work

#PlanScorePrice (US)Price (DE)
#1 Claude MaxAnthropic 94.5 $100.00 –
#2 Google AI Ultra (5x)Google 94.0 $99.99 99.99 €
#3 ChatGPT Pro 100OpenAI 90.6 $100.00 103.00 €
#4 The X Subscription SuperxAI 81.0 $100.00 105.00 €
#5 Microsoft 365 ProMicrosoft 80.5 $99.99 110.00 €

All criteria for advanced plans (about 100 a month) · Best AI subscription for coding

#PlanScorePrice (US)Price (DE)
#1 Claude MaxAnthropic 94.5 $100.00 –
#2 Google AI Ultra (5x)Google 94.0 $99.99 99.99 €
#3 ChatGPT Pro 100OpenAI 90.6 $100.00 103.00 €
#4 Microsoft 365 ProMicrosoft 90.0 $99.99 110.00 €
#5 The X Subscription SuperxAI 81.0 $100.00 105.00 €

All criteria for advanced plans (about 100 a month) · Best AI subscription for research

Free plans

# Plan Score Price (US) Price (DE) Best included model (Text Arena) 40 %Features 20 %Usage allowance (editorial, 0–10) 20 %
#1 Gemini FreeGoogle 76.0 $0.00 0.00 € Gemini 3.5 Flash-Lite (1,455)78/100 7.5 of 1075/100 5/1050/100
#2 Claude FreeAnthropic 71.1 $0.00 – Claude Sonnet 5.5 (1,476)85/100 4.5 of 1045/100 4/1040/100
#3 ChatGPT FreeOpenAI 71.1 $0.00 0.00 € GPT-6 Luna (1,442)73/100 5 of 1050/100 6/1060/100
#4 Mistral Vibe FreeMistral AI 67.0 $0.00 – Mistral Large 4 (1,434), limited access70/100 5.5 of 1055/100 4/1040/100
#5 Copilot FreeMicrosoft 66.7 $0.00 0.00 € No data 4.5 of 950/100 5/1050/100
#6 The X Subscription FreexAI 66.2 $0.00 – Grok 4.7 (1,441), limited access73/100 4.5 of 856/100 3/1030/100
#7 Perplexity FreePerplexity 65.0 $0.00 – No data 4.5 of 1045/100 5/1050/100

#1 Gemini Free Google

Free Gemini app with a Google Account: 3.5 Flash-Lite under standard compute-based limits (32k-token context window), image generation with Nano Banana 2, Deep Research, Gemini Live, Canvas and Gems, plus 15 GB of Google cloud storage.

Price
$0.00 per month (US, with a Google Account) 0.00 € per month (DE, mit einem Google-Konto)
Models
Gemini 3.5 Flash-Lite full Free plan card: "Access to our Flash-Lite model". The plans page no longer names the version; 3.5 Flash-Lite is the current Flash-Lite model in the Gemini API (see open_questions). Gemini 3.6 Flash limited Release notes (2026-07-21) bring 3.6 Flash "to all Gemini app users globally"; the plans page lists it only from Google AI Plus. Standard limits. Gemini 3.1 Pro limited Release notes (2026-02-19): "rolling out globally to the Gemini app, with higher limits for users with Google AI Pro and Ultra plans". Standard limits. Nano Banana 2 full Image generation and editing model, labelled "Free" on the image generation page. Nano Banana Pro limited Only in Google Flow: "limited access to Nano Banana Pro".
Limits
Gemini app usage: Standard limitsLimit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.Context window: 32k tokensHigh demand: For users without a Google AI plan, some features that require more compute resources, like Deep Research, may be unavailable during periods of high demand.File uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Audio and video upload length: Total video length can be up to 5 minutes. Total audio length can be up to 10 minutes.Jules (free plan): Daily Tasks (rolling 24 hours): 15; Concurrent Tasks: 3Cloud storage: 15 GB of cloud storage across Gmail, Drive, and Photos
Usage score
5/10 · Standard limits that refresh every 5 hours up to a weekly cap, with no numbers published, and a 32k-token context window.
  • Web search
  • Deep research (Listed for the free plan, but may be unavailable without a Google AI plan during periods of high demand; Pro and Ultra get higher report limits.)
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent (Jules free plan: 15 daily tasks (rolling 24 hours) and 3 concurrent tasks (Jules docs); Google Antigravity is listed as "No cost to start" with "Basic weekly rate limits".)
  • Browser or computer agent (Only screen automation (beta) in the Android app on Pixel 10 / Pixel 11 series or Galaxy S26 / Z Flip 8 / Z Fold 8 / Tab S12, in India, Japan, South Korea, the UK or the US, English, Japanese and Korean, select apps. Tasks in Gemini and Chrome auto browse need Pro or Ultra.)
  • Custom assistants or projects
  • Memory across chats

#2 Claude Free Anthropic

Claude's no-cost plan for everyday questions with the Sonnet and Haiku models, web search, memory, file creation and voice mode, under a usage limit that resets every five hours; the pricing page also lists a weekly limit.

Price
$0.00 per month (US, "Free for everyone")
Models
Claude Sonnet 5.5 full The pricing table lists the Sonnet family for Free without a version number; Sonnet 5.5 is the current Sonnet model. Claude Haiku 5.5 full The pricing table lists the Haiku family for Free without a version number; Haiku 5.5 is the current Haiku model.
Limits
Session limit: "there is a session-based usage limit that will reset every five hours"; "the number of messages you can send will vary based on demand"Overall usage: "Every plan has usage limits. Most accounts have a limit that resets on a rolling five-hour session window, and paid plans add weekly limits on top."; "there's no fixed message count"; "Free covers everyday questions." (no absolute number published)Weekly limit: Pricing card: "Limited usage every 5 hours and every week"; comparison table "Limit every week": "Limited" (no number published; the FAQ and the Free help article name weekly limits only for paid plans)Projects: "Free users can create a maximum of five projects."Opus and Fable models: Not included (pricing table: Opus "No", Fable "No")Usage credits: Not available (pricing table: "No")
Usage score
4/10 · A session limit that resets every five hours; no number is published and the message count varies with demand.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects (Projects with project instructions; Free users can create up to five projects.)
  • Memory across chats

#3 ChatGPT Free OpenAI

Free entry tier with unlimited everyday text chats on GPT-6 (powered by GPT-6 Luna) and capped access to uploads, image creation, voice, deep research, Codex and ChatGPT Work on desktop. Ads may appear in certain countries; the optional Ads-Free mode there has lower limits.

Price
$0.00 per month (US) 0.00 € per month (DE, Shown as '0 €' on the German pricing page)
Models
GPT-6 Luna full Powers GPT-6 in Chat for Free (rollout to Free and Go from Oct 8, 2026); pricing page: 'Yes'. Also in Codex: 'GPT-6 Luna at Standard speed in the desktop app, subject to rollout'. GPT-5.6 Luna limited Help center: Free users can use Think for harder questions; 'Think uses GPT-5.6 Luna, not GPT-5.6 Sol'. The pricing page comparison marks GPT-5.6 Luna as not included for Free. GPT-5.6 Terra limited Pricing page: 'Limited access in Work and Codex on desktop'; help center: 'Codex: Terra for Free and Go'. GPT-5 Thinking Mini full Pricing page comparison: 'Yes'.
Limits
Everyday text chats: Unlimited text chats subject to abuse guardrailsAds-Free option (regions with ads): Switching Free to Ads-Free 'removes ads, but comes with lower usage limits and reduced feature access (for example, fewer messages and no access to some tools like image generation or deep research)' (no numbers published)Tools (uploads, images, voice, data analysis): File uploads, image generation, voice, data analysis, and other tools have separate usage limits (no numbers published)Voice in Chat (Live, rolling 24-hour period): Limited access to GPT-Live-1.1 Quip. Limits may change.Deep research: Limited deep research; usage varies by plan (no number published)Codex and ChatGPT Work: Limited Codex access; Limited ChatGPT Work access on desktop (no number published)GPT Instant total context window: 27KGPT Instant input maximum: ~12 pages of textGPT Reasoning total context window: VariesLibrary storage: Use Library with 500 MB of storage
Usage score
6/10 · Unlimited text chats subject to abuse guardrails; uploads, images, voice and deep research have separate caps without published numbers.
  • Web search
  • Deep research (Pricing page: 'Limited deep research'; no task count published.)
  • Image generation (Pricing page: 'Limited image creation'; image generation has a separate usage limit (no number published).)
  • Video generation
  • Voice mode (Pricing page: 'Limited voice chats'; Voice help: 'Limited access to GPT-Live-1.1 Quip'.)
  • File uploads (Pricing page: 'Limited messages with uploads'.)
  • Coding agent (Codex is included on Free with limited access; Codex Cloud is not included with Free.)
  • Browser or computer agent (ChatGPT Work (agent for multi-step tasks) is 'Limited (desktop app)'; the Work cloud browser excludes Free and Go. The former ChatGPT agent mode is no longer available.)
  • Custom assistants or projects (Projects and shared projects are included and existing GPTs/plugins can be used; creating and sharing plugins is not included on Free.)
  • Memory across chats (Pricing page: 'Limited memory and context'.)

#4 Mistral Vibe Free Mistral AI

The free tier of Mistral's assistant Le Chat (now renamed Mistral Vibe) for quick answers and simple everyday tasks on web and mobile, with limited messages, web searches and coding sessions, basic image generation and 100+ connectors.

Price
$0.00 per month (US, Free plan; the pricing page shows no price for it)
Models
Mistral Medium 3.5 limited Vibe Code Web uses Mistral Medium 3.5 (Free: 2 sessions per day, where access is enabled). The model answering Free chats is not named; the pricing FAQ says Free includes 'our latest models'. Mistral Large 4 limited The Oct 29, 2026 release says 'You can also try the new model in Vibe Code'; no plan restriction is stated. On Free, Vibe Code is limited (CLI/IDE trial, 2 Vibe Code Web sessions a day where enabled). Not announced for Work or chat; the model card still says Public Preview. Black Forest Labs image models limited Work uses Black Forest Labs models for image generation (model not named); Free gets 'Basic image generation and editing' with daily limits by plan. Voxtral (speech-to-text for Voice mode) full Voice mode transcription is available on all plans. The docs page now omits the model name ('powered by , a speech-to-text model we developed'); the Vibe product page says voice input is 'powered by Voxtral', and Mistral's speech-to-text models are all Voxtral.
Limits
Messages and web searches: Limited messages and web searches. (no number published)Coding: Limited coding sessions. (no number published)Vibe Code Web sessions: Free users: 2 sessions per day, where access is enabledIncluded monthly usage (Studio, API, Vibe Code): Each Mistral plan includes monthly usage that is shared across Studio, the API, and Vibe Code. Usage consumes this monthly allowance first. (no amount published for Free)Image generation: Basic image generation and editing. (no number published)Libraries and document upload: LimitedLibrary uploads: Up to 100 files can be uploaded at once per Library. Maximum file size: 100 MB per file. Each document uses processing capacity based on its size; the monthly allowance is shared across all your Libraries (allowance size not published).Task scheduling: Up to 5When rate limits are reached: If you reach these limits, your access to the Mistral AI Products will be temporarily suspended until the limits reset or you upgrade your subscription.
Usage score
4/10 · Limited messages and web searches with no numbers published, and 2 Vibe Code Web sessions a day where enabled.
  • Web search
  • Deep research (Deep Research is a built-in Skill in Work. The pricing FAQ says Pro adds 'more extended thinking and deep research reports'; the Free allowance is not published.)
  • Image generation ('Basic image generation and editing.' with daily generation limits by plan; Pro allows up to 40x more image generations.)
  • Video generation
  • Voice mode (Voice mode is speech-to-text input (Voxtral): speech is transcribed into a prompt you can send. No spoken replies are documented.)
  • File uploads ('Libraries and document upload' is 'Limited' on Free (no size published); Libraries take up to 100 files per upload and 100 MB per file, with a monthly processing allowance (size not published). Docs say Files and Canvas are available on all plans with plan-based usage limits.)
  • Coding agent (Vibe Code (CLI, IDE plugin, web) with 'Limited coding sessions'; Vibe Code Web: 2 sessions a day where enabled. Draws on the plan's monthly usage shared with Studio and the API. Mistral Large 4 can be tried in Vibe Code (Oct 29, 2026). Hooks, remote agents and Teleport need Pro.)
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#5 Copilot Free Microsoft

The free Copilot app offers chat (GPT-5 Smart mode, Think Deeper), web search, image creation and file uploads, subject to available capacity and limits. Signing in with a personal Microsoft account adds chat history, longer conversations, voice and memory; agents and higher limits require a Microsoft 365 subscription.

Price
$0.00 per month (US, Pricing page FAQ 'Is Copilot free?': 'Yes. Microsoft offers a limited free version of Copilot subject to available capacity and limits.') 0.00 € per month (DE, German pricing page FAQ: 'Microsoft bietet eine eingeschränkte kostenlose Version von Copilot an, vorbehaltlich verfügbarer Kapazitäten und Nutzungslimits.')
Models
GPT-5 full Smart mode: 'Smart mode is powered by GPT-5' and 'Smart mode is available to all Copilot users'. Free use is subject to available capacity and limits. GPT-5 is not in the OpenAI catalog. Think Deeper (OpenAI reasoning models, not named) full 'Think Deeper is powered by the latest reasoning models from OpenAI.' 'Use of Think Deeper in Copilot is not limited'; paying Microsoft 365 users get priority access when capacity is limited. MAI-Image-2 limited Microsoft AI, Mar 19, 2026: 'MAI-Image-2 is beginning to roll out on Copilot and Bing Image Creator.' Free image creation is subject to capacity and limits; the model serving a given request is not stated. MAI-Image-2-Efficient limited Microsoft AI, Apr 14, 2026: 'It's also rolling out across Copilot and Bing, with more surfaces like PowerPoint coming soon.' The model serving a given request is not stated.
Limits
Free version: Microsoft offers a limited free version of Copilot subject to available capacity and limits. When you hit a limit, you can choose to upgrade to a paid option or come back later.Included for free: You can continue to chat with Copilot, create images, upload files and more for free, subject to available capacity and limitsHigher limits and agents: For higher limits to chat and create, use agents, or tackle complex multi-step tasks, purchase a Microsoft 365 subscription.Overall usage and AI image generation (compare page): Limited usage with a free Microsoft accountThink Deeper: Use of Think Deeper in Copilot is not limited; Microsoft 365 Personal, Family, and Premium users have priority access whenever capacity is limitedSign-in: Sign-in is optional; signing in with a personal account gives you access to chat history, image creation, longer conversations, voice interactionsUsage limits (numbers): not published
Usage score
5/10 · A limited free version subject to available capacity; no numbers published, while Think Deeper is described as not limited.
  • Web search
  • Deep research
  • Image generation (Free image creation is 'subject to available capacity and limits'; paid plans advertise 'Higher usage than free for AI image creation in Copilot'. No numbers published.)
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#6 The X Subscription Free xAI

Free tier of The X Subscription, the single plan xAI sells for Grok, Cursor and X since October 12, 2026. It gives limited daily access to Grok Chat, Voice, Imagine (images only) and Grok Build, a starter allowance for Cursor's coding agents and standard X, without video generation, Grok Bot or Grok API access.

Price
$0.00 per month (US)
Models
Grok Chat (model not named) limited x.ai/pricing lists Chat as 'Limited' on Free; no official page names the model used for free Chat. Grok 4.7 limited Via Grok Build, which is 'Limited' on Free (x.ai/pricing); x.ai/build says Grok Build is 'now powered by Grok 4.7' and 'Available to try for Free'. Grok Imagine (images) limited Image generation (Imagine) is 'Limited' on Free; video generation is not included. The Imagine model per plan is not stated.
Limits
Included usage: LimitedAI features: Limited Chat requests; Limited agent requestsChat, Image generation (Imagine), Grok Build: LimitedWhat's included: Daily access to Grok, a starter allowance for Cursor's coding agents, and standard XFree-tier Chat and Voice limits: Grok's free tier limits on Chat and Voice reset on their own schedule (no numbers published)Usage limits: not published (no absolute numbers)
Usage score
3/10 · Limited chat and agent requests with no numbers published and only a starter allowance for Cursor's coding agents.
  • Web search
  • Deep research
  • Image generation (Image generation (Imagine) is 'Limited' on Free (x.ai/pricing).)
  • Video generation
  • Voice mode (Docs FAQ: 'You will still have access to Grok's free tier limits on Chat and Voice'; the announcement describes Free as 'Daily access to Grok'. No numbers are published.)
  • File uploads
  • Coding agent (Grok Build is 'Limited'; Cursor Agent is included with 'a starter allowance for Cursor's coding agents'.)
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#7 Perplexity Free Perplexity

Free tier for light use: unlimited basic search with citations and automatic model choice, plus 3 Pro Searches per day (applied automatically) and one Research query per month. No model picker, no Projects and no video generation; Perplexity Computer only as a one-time trial of free turns on some accounts.

Price
$0.00 per month (US, Called "Standard (Free)" in the help center.)
Models
Best (automatic model selection) full "We pick the best model to use for your query"; Free users cannot choose an advanced Search model manually (pricing page: "Access to basic AI models").
Limits
Basic searches: Practically unlimited basic searchesPro Searches: 3/dayPro Search use: Pro Search is no longer a separate toggle, so there’s nothing to turn on. [...] Perplexity uses Pro Search automatically when it’s included in your plan, up to your plan’s limit.Research Queries: 1/monthResearch limit reset: Research limits reset on a rolling basis rather than on a fixed calendar date. [...] access returns automatically as older Research queries roll out of the usage window.Browser Agent (Comet Assistant) Queries: NoFile and App Creation: NoFile Uploads (sessions): LimitedFile upload allowance: File uploads are not unlimited on Free, Pro, or Max plans. Each plan includes an allowance of file uploads that resets on a rolling weekly basis, and Max includes a higher allowance than Pro. Images you attach or paste into a search count toward the same allowance. [...] There is no fixed weekly reset day.File size: File size is limited to 40 MB for all file types.Image generation: Free Plan: Get access to limited image generations.Get answers from the web (pricing table): LimitedPerplexity Computer: Some free accounts include a limited number of free Computer turns. A turn is one message you send to Computer and Computer’s reply to it. Free turns are a one-time allowance and don’t use credits. You can only use credits with an active paid plan.
Usage score
5/10 · Practically unlimited basic searches, but only 3 Pro Searches a day and 1 research query a month.
  • Web search
  • Deep research (Research queries: 1/month ("Free users get limited access to Research").)
  • Image generation (Image help article: "Free Plan: Get access to limited image generations." The plan comparison table lists Image Generation as "No" for Free (conflict).)
  • Video generation
  • Voice mode
  • File uploads (Basic file uploads (limited), rolling weekly allowance with no fixed reset day (images attached or pasted into a search count too), 40 MB per file.)
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

Entry plans (under 15 a month)

# Plan Score Price (US) Price (DE) Best included model (Text Arena) 40 %Features 20 %Usage allowance (editorial, 0–10) 20 %
#1 Google AI PlusGoogle 82.1 $4.99 4.99 € Gemini 3.6 Flash (1,483)88/100 8.5 of 1085/100 5/1050/100
#2 Google AI Plus (2 TB)Google 77.6 – 9.99 € Gemini 3.6 Flash (1,483)88/100 8.5 of 1085/100 5/1050/100
#3 ChatGPT GoOpenAI 68.6 $8.00 8.00 € GPT-6 Luna (1,442)73/100 7.5 of 1075/100 6/1060/100
#4 The X Subscription PlusxAI 64.5 $8.00 8.00 € Grok 4.7 (1,441), limited access73/100 6 of 875/100 4/1040/100
#5 Mistral Vibe ProMistral AI 62.5 $14.99 17.84 € Mistral Medium 3.5 (1,426)67/100 7.5 of 1075/100 7/1070/100
#6 Microsoft 365 PersonalMicrosoft 53.3 $9.99 10.00 € No data 6 of 1060/100 5/1050/100
#7 Microsoft 365 FamilyMicrosoft 49.5 $12.99 13.00 € No data 6 of 1060/100 5/1050/100

#1 Google AI Plus Google

Entry paid plan with twice the free usage limits: 3.6 Flash and, per the help center, the Pro model (128k-token context window), Nano Banana 2.1 images, limited Omni Flash video generation, 200 Google Flow credits per month, Gemini in Gmail and 400 GB of Google One storage. Daily Brief is rolling out to subscribers in the US.

Price
$4.99 per month (US, 400 GB storage) 4.99 € per month (DE, 400 GB storage)
Models
Gemini 3.6 Flash full Google One AI plans page: "Access to Gemini 3.6 Flash and Deep Research in Gemini"; the Gemini plans page says "Access to our Flash model". Gemini 3.1 Pro limited Help center model page: Google AI Plus chat models are Flash-Lite, Flash and Pro; 128k-token context window. Conflicting: the Google One comparison table marks "Access to our Pro models" as unavailable for Plus (checked 2026-11-01) and the plan cards list only the Flash model (see open_questions). Gemini 3.5 Flash-Lite full Included from the free plan; chats with a Google AI plan continue with Flash-Lite once the usage limit is reached. Nano Banana 2.1 full "Google AI Pro, Plus, Ultra users have access to our best image model, Nano Banana 2.1" Nano Banana Pro limited Only for regenerating images in the Gemini app via "Redo with Pro". Gemini Omni 1.1 Flash (video) limited "limited access to Gemini Omni Flash" in the Gemini app; 10-second videos, 18+.
Limits
Gemini app usage: 2x higher than standard limitsGemini app usage (Gemini plans page): Get double the usage limits than Free and access to more features like video generation.Access to Pro models (Google One comparison): Not included (marked unavailable for Google AI Plus)Limit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.When the limit is reached: If you have a Google AI plan and reach your usage limit, your chat will continue with Flash-Lite.Context window: 128k tokensFile uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Google Flow credits: 200/moGoogle Flow Music credits: 3,000/moGoogle Antigravity agent requests (Google One comparison): LimitedAI credits: AI credits aren't available for purchase to Google AI Plus members.Cloud storage: 400 GB of cloud storage across Gmail, Drive, and PhotosBilling options: You can choose between a monthly or annual membership plan for Google AI Plus.
Usage score
5/10 · Twice the standard Gemini limits, refreshing every 5 hours up to a weekly cap, a 128k-token context and 200 Flow credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation ("limited access to Gemini Omni Flash" in the Gemini app; 200 Google Flow credits per month with access to Omni Flash. 10-second videos, 18+.)
  • Voice mode
  • File uploads
  • Coding agent (Google Antigravity "Agent requests" are "Limited" in the Google One comparison (lowest paid level). Jules is no longer listed for Plus on the plans pages or the Plus benefits help page; the Jules docs list only the free plan (15 daily tasks), Pro and Ultra.)
  • Browser or computer agent (Only screen automation (beta) on select Android phones and tablets in India, Japan, South Korea, the UK or the US; Tasks in Gemini and Chrome auto browse require Google AI Pro or Ultra.)
  • Custom assistants or projects
  • Memory across chats

#2 Google AI Plus (2 TB) Google

Storage variant of Google AI Plus: the same Gemini features and limits with 2 TB instead of 400 GB of Google One storage, plus longer calls, noise cancellation and recording in Google Meet. Google One lists it at 9.99 EUR a month or 99.99 EUR a year in Germany; no US price was confirmed on an official page.

Price
9.99 € per month (DE, Google AI Plus (2 TB)) 99.99 € per year (DE, Google AI Plus (2 TB), annual billing)
Models
Gemini 3.6 Flash full Google One AI plans page: "Access to Gemini 3.6 Flash and Deep Research in Gemini"; the Gemini plans page says "Access to our Flash model". Gemini 3.1 Pro limited Help center model page: Google AI Plus chat models are Flash-Lite, Flash and Pro; 128k-token context window. Conflicting: the Google One comparison table marks "Access to our Pro models" as unavailable for Plus (checked 2026-11-01) and the plan cards list only the Flash model (see open_questions). Gemini 3.5 Flash-Lite full Included from the free plan; chats with a Google AI plan continue with Flash-Lite once the usage limit is reached. Nano Banana 2.1 full "Google AI Pro, Plus, Ultra users have access to our best image model, Nano Banana 2.1" Nano Banana Pro limited Only for regenerating images in the Gemini app via "Redo with Pro". Gemini Omni 1.1 Flash (video) limited "limited access to Gemini Omni Flash" in the Gemini app; 10-second videos, 18+.
Limits
Gemini app usage: 2x higher than standard limitsGemini app usage (Gemini plans page): Get double the usage limits than Free and access to more features like video generation.Access to Pro models (Google One comparison): Not included (marked unavailable for Google AI Plus)Limit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.When the limit is reached: If you have a Google AI plan and reach your usage limit, your chat will continue with Flash-Lite.Context window: 128k tokensFile uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Google Flow credits: 200/moGoogle Flow Music credits: 3,000/moGoogle Antigravity agent requests (Google One comparison): LimitedAI credits: AI credits aren't available for purchase to Google AI Plus members.Cloud storage: Choose from 400 GB or 2 TB of storage.Billing options: You can choose between a monthly or annual membership plan for Google AI Plus.2 TB option: The Google AI Plus 2 TB plan includes all benefits of the 400 GB plan.Google Meet (2 TB option): Get longer calls, noise cancellation, and conversation recording in Google Meet.
Usage score
5/10 · Twice the standard Gemini limits, refreshing every 5 hours up to a weekly cap, a 128k-token context and 200 Flow credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation ("limited access to Gemini Omni Flash" in the Gemini app; 200 Google Flow credits per month with access to Omni Flash. 10-second videos, 18+.)
  • Voice mode
  • File uploads
  • Coding agent (Google Antigravity "Agent requests" are "Limited" in the Google One comparison (lowest paid level). Jules is no longer listed for Plus on the plans pages or the Plus benefits help page; the Jules docs list only the free plan (15 daily tasks), Pro and Ultra.)
  • Browser or computer agent (Only screen automation (beta) on select Android phones and tablets in India, Japan, South Korea, the UK or the US; Tasks in Gemini and Chrome auto browse require Google AI Pro or Ultra.)
  • Custom assistants or projects
  • Memory across chats

#3 ChatGPT Go OpenAI

Low-cost plan for longer conversations: everything in Free plus higher limits for tools, uploads, image creation and voice, longer memory and scheduled tasks. Still runs on GPT-6 Luna in Chat, has only limited Codex and desktop Work access, and ads are being tested with eligible Go users in select regions.

Price
$8.00 per month (US, Billed monthly; no annual plan) 8.00 € per month (DE, Shown as '8 €' / month on the German pricing page)
Models
GPT-6 Luna full Powers GPT-6 in Chat for Go (rollout to Free and Go from Oct 8, 2026); pricing page: 'Yes'. Also in Codex: 'GPT-6 Luna at Standard speed in the desktop app, subject to rollout'. GPT-5.6 Luna limited Go help: 'ChatGPT Go users can access Think ... Think uses GPT-5.6 Luna. ChatGPT Go does not include GPT-5.6 Sol.' The pricing page comparison marks GPT-5.6 Luna as not included for Go. GPT-5.6 Terra limited Pricing page: 'Limited access in Work and Codex on desktop'; help center: 'Codex: Terra for Free and Go'. GPT-5 Thinking Mini full Pricing page comparison: 'Yes'.
Limits
Everyday text chats: Unlimited text chats subject to abuse guardrailsTools (images, uploads, data analysis): ChatGPT Go includes higher limits for eligible features such as image generation, file uploads, and data analysis (no numbers published)Voice in Chat (Live, rolling 24-hour period): 3 hours with GPT-Live-1.1 QuipDeep research: Limited (pricing page comparison); usage varies by plan (no number published)Codex and ChatGPT Work: Codex: Limited; ChatGPT Work: Limited (desktop app) (no number published)GPT Instant total context window: 54KGPT Instant input maximum: ~40 pages of textGPT Reasoning total context window: 256KGPT Reasoning input maximum: ~320 pages of text
Usage score
6/10 · Unlimited text chats subject to abuse guardrails, higher tool limits than Free and 3 hours of voice a day; deep research and Codex are limited.
  • Web search
  • Deep research (Pricing page comparison: 'Limited'; no task count published.)
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent (Codex is included on Go with limited access; Codex Cloud is not included with Go.)
  • Browser or computer agent (ChatGPT Work is 'Limited (desktop app)'; the Work cloud browser and event-triggered Work tasks exclude Free and Go.)
  • Custom assistants or projects
  • Memory across chats

#4 The X Subscription Plus xAI

Entry paid tier of The X Subscription, recommended by xAI for social creators with limited AI usage. It adds a shared weekly AI usage pool across Grok and Cursor for Grok Chat, Voice, Imagine (images) and Build, Grok API access (billed separately at standard API rates, outside the pool), plus a checkmark, creator tools and fewer ads on X. Video generation and Grok Bot are not included.

Price
$8.00 per month (US, excl. taxes ('Prices shown do not include applicable taxes or fees (e.g. VAT)')) $80.00 per year (US, billed annually ($7/mo); list price excl. sales tax) 8.00 € per month (DE, excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren'), as shown on grok.com/subscription to a signed-out visitor in Germany) 80.00 € per year (DE, 80 €/Jahr (7 €/Monat), as shown on grok.com/subscription in Germany; excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren'))
Models
Grok Chat (model not named) full Plus includes 'Grok Chat, Voice, Imagine, and Build'; 'Latest Grok models' is listed only from Premium up, so the Chat models on Plus are not named. Grok 4.7 limited Confirmed via Grok Build, which Plus includes and which is 'now powered by Grok 4.7'; whether Plus gets Grok 4.7 in Chat is not stated. Grok Imagine (images) full Image generation (Imagine) included; video generation is not included on Plus.
Limits
Included usage: Shared AI usage pool across all productsUsage tier: BaseUsage pool: Every tier includes a pool of AI usage shared across Grok and Cursor. Within Grok, the pool covers Chat, Imagine, Voice, and Bot. Higher tiers include larger pools, and usage costs vary by product.Grok API: Paid plans include Grok API access. API usage is billed separately at standard model rates and doesn't draw from your plan's usage pool.Refresh: Your allowance refreshes weekly, so one heavy day never wipes out your whole month.Beyond included usage: If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh.Usage limits: not published (no absolute numbers)
Usage score
4/10 · The base usage pool shared across Grok, Cursor and X, refreshing weekly; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#5 Mistral Vibe Pro Mistral AI

Mistral's paid personal plan for Le Chat (now Mistral Vibe), aimed at complex tasks, deeper reasoning and all-day coding. Compared with Free it adds up to 6x the messages, 5x the web searches and 40x the image generations, spreadsheets, slides, Canvas, Code Interpreter, remote coding agents, $25.5/mo in API credits and chat and email support.

Price
$14.99 per month (US, excl. taxes (pricing page: 'Excluding taxes')) 17.84 € per month (DE, incl. taxes, shown with EUR and country Germany selected; EUR list price 14.99 excl. taxes)
Models
Mistral Medium 3.5 full Announced as the default model in Le Chat, powering Work mode and remote coding agents on Pro, Team and Enterprise; Vibe Code Web uses Mistral Medium 3.5. Mistral Large 4 preview The Oct 29, 2026 release says 'You can also try the new model in Vibe Code'; no plan restriction is stated. Not announced for Work or chat; the model card still says Public Preview, and the Vibe Code Web limits page still says it 'Uses Mistral Medium 3.5'. Black Forest Labs image models full Work uses Black Forest Labs models for image generation (model not named); Pro: 'Up to 40x free' image generations. Voxtral (speech-to-text for Voice mode) full Voice mode transcription is available on all plans. The docs page now omits the model name ('powered by , a speech-to-text model we developed'); the Vibe product page says voice input is 'powered by Voxtral', and Mistral's speech-to-text models are all Voxtral.
Limits
Overall usage: Subject to fair usage limits and Mistral's Terms of Service.Messages: Up to 6x free (no absolute number published)Web searches: Up to 5x free (no absolute number published)Image generation: Up to 40x free (no absolute number published)Flash answers: 150 / day on ProLibraries and document upload: Up to 15GBLibrary uploads: Up to 100 files can be uploaded at once per Library. Maximum file size: 100 MB per file. Each document uses processing capacity based on its size; the monthly allowance is shared across all your Libraries (allowance size not published).Task scheduling: Unlimited (Subject to fair usage limits and Mistral's Terms of Service.)Project folders: Up to 1,000Vibe Code Web sessions: Paid users: 100 sessions per dayIncluded monthly usage (Studio, API, Vibe Code): Each Mistral plan includes monthly usage that is shared across Studio, the API, and Vibe Code. Usage consumes this monthly allowance first. (no amount published; the pricing page does not say whether the '$25.5 /mo in API credits' is this allowance)Included API credits: $25.5 /mo in API credits.Beyond plan limits: Extend usage with PAYG credits: Keep going past your plan limit by paying only for what you use beyond it at API rate.When included monthly usage is exhausted: If pay-as-you-go is enabled, additional usage draws from your prepaid credits. If it isn't enabled, usage can stop until the next billing period or until an admin changes the settings.When rate limits are reached: If you reach these limits, your access to the Mistral AI Products will be temporarily suspended until the limits reset or you upgrade your subscription.
Usage score
7/10 · Up to 6x the free messages and 40x the free image generations, 100 Vibe Code Web sessions a day and $25.5 of API credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode (Voice mode is speech-to-text input (Voxtral): speech is transcribed into a prompt you can send. No spoken replies are documented.)
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#6 Microsoft 365 Personal Microsoft

Single-user Microsoft 365 plan (sold as Microsoft 365 Single in Germany) that bundles Word, Excel, PowerPoint and Outlook with Copilot, 1 TB of OneDrive storage, AI credits and higher Copilot usage than the free version. Researcher, Analyst and Cowork are not included.

Price
$9.99 per month (US, '$9.99 /month'; 'Try free for 1 month'; renews automatically unless canceled.) $99.99 per year (US, '$99.99 /year') 10.00 € per month (DE, Sold as 'Microsoft 365 Single' in Germany; shown as '10,00 € /Monat'; '1 Monat lang kostenlos testen'. VAT is not mentioned on the page.) 99.00 € per year (DE, Microsoft 365 Single, shown as '99,00 € /Jahr'.)
Models
GPT-5 full Smart mode: 'Smart mode is powered by GPT-5'. GPT-5 is not in the OpenAI catalog. Think Deeper (OpenAI reasoning models, not named) full 'Think Deeper is powered by the latest reasoning models from OpenAI.' Personal users 'have priority access whenever capacity is limited'. MAI-Image-2 full Microsoft AI, Mar 19, 2026: 'MAI-Image-2 is beginning to roll out on Copilot and Bing Image Creator.' Image generation: 'Standard usage' on Personal. MAI-Image-2-Efficient full Microsoft AI, Apr 14, 2026: 'It's also rolling out across Copilot and Bing'. The model serving a given request is not stated.
Limits
Chat: Extensive usageDeep reasoning: Extensive usagePages: Extensive usageAudio Overview: Standard usageVoice: Standard usageVision: Standard usageImage gen: Standard usageResearcher, Analyst, Cowork (in preview): Not availablePriority access: YesAI credits: Microsoft 365 Personal and Family include AI credits and feature limits (monthly allotment; no number published)When a limit is reached: If you reach a feature limit or use all of your AI credits, that feature may be unavailable until your allowance resets.Overall usage (pricing page): Advanced Copilot chat with higher usage than free; Higher usage than free for AI image creation in Copilot; Higher usage than free for visual AI assistance from your screen or camera; Higher usage than free for AI-generated audio (English only)Copilot in Office apps: Word, Excel, PowerPoint, and Outlook desktop apps with Microsoft Copilot; Use Copilot in select apps with work files in a secure wayUsers, devices, storage: For 1 person; Sign into 5 devices at once; 1 TB of secure cloud storageAI features: AI features only available to subscription owner and cannot be shared; usage limits apply.
Usage score
5/10 · Extensive usage for chat and deep reasoning in Microsoft's table, standard usage for voice, vision and images; Researcher is not included.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#7 Microsoft 365 Family Microsoft

Microsoft 365 for up to six people with 2 TB of shared OneDrive storage and the Office apps; its Copilot features and AI credits match Personal but are available only to the subscription owner.

Price
$12.99 per month (US, '$12.99 /month'; 'Try free for 1 month'; renews automatically unless canceled.) $129.99 per year (US, '$129.99 /year') 13.00 € per month (DE, Shown as '13,00 € /Monat'; '1 Monat lang kostenlos testen'. VAT is not mentioned on the page.) 129.00 € per year (DE, Shown as '129,00 € /Jahr'.)
Models
GPT-5 full Smart mode: 'Smart mode is powered by GPT-5'. Copilot features are for the subscription owner only. GPT-5 is not in the OpenAI catalog. Think Deeper (OpenAI reasoning models, not named) full 'Think Deeper is powered by the latest reasoning models from OpenAI.' Family users 'have priority access whenever capacity is limited'. MAI-Image-2 full Microsoft AI, Mar 19, 2026: 'MAI-Image-2 is beginning to roll out on Copilot and Bing Image Creator.' Image generation: 'Standard usage' on Family. MAI-Image-2-Efficient full Microsoft AI, Apr 14, 2026: 'It's also rolling out across Copilot and Bing'. The model serving a given request is not stated.
Limits
Chat: Extensive usageDeep reasoning: Extensive usagePages: Extensive usageAudio Overview: Standard usageVoice: Standard usageVision: Standard usageImage gen: Standard usageResearcher, Analyst, Cowork (in preview): Not availablePriority access: YesAI credits: Microsoft 365 Personal and Family include AI credits and feature limits (monthly allotment; no number published)When a limit is reached: If you reach a feature limit or use all of your AI credits, that feature may be unavailable until your allowance resets.Overall usage (pricing page): Advanced Copilot chat with higher usage than free; Higher usage than free for AI image creation in Copilot; Higher usage than free for visual AI assistance from your screen or camera; Higher usage than free for AI-generated audio (English only)Copilot in Office apps: Word, Excel, PowerPoint, and Outlook desktop apps with Microsoft Copilot (Everything in Microsoft 365 Personal)Users, devices, storage: For 1 to 6 people (AI for subscription owner only); Each person can use on up to 5 devices simultaneously; 2 TB of shared, secure cloud storageAI features: AI features only available to subscription owner and cannot be shared; usage limits apply.
Usage score
5/10 · Extensive usage for chat and deep reasoning in Microsoft's table, standard usage for voice, vision and images; AI is for the owner only.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

Standard plans (about 20 a month)

# Plan Score Price (US) Price (DE) Best included model (Text Arena) 40 %Features 20 %Usage allowance (editorial, 0–10) 20 %
#1 Google AI ProGoogle 94.0 $19.99 21.99 € Gemini 4 Argon (1,517)100/100 10 of 10100/100 7/1070/100
#2 ChatGPT PlusOpenAI 86.6 $20.00 23.00 € GPT-6.1 Sol (1,480)87/100 9 of 1090/100 7/1070/100
#3 Claude ProAnthropic 86.5 $20.00 – Claude Opus 5.5 (1,507)96/100 8 of 1080/100 6/1060/100
#4 Perplexity ProPerplexity 83.9 $20.00 – Gemini 3.8 Flash (1,496)92/100 8.5 of 1085/100 5/1050/100
#5 Microsoft 365 PremiumMicrosoft 76.7 $19.99 22.00 € No data 7 of 1070/100 6/1060/100
#6 The X Subscription PremiumxAI 74.3 $30.00 31.00 € Grok 4.7 (1,441)73/100 9 of 9100/100 6/1060/100

#1 Google AI Pro Google

Standard plan with 4x the free usage limits: Gemini 4 Argon (rolling out since 13 October 2026), the Flash and Pro models (3.1 Pro) with a 1M-token context window, higher Deep Research limits, Omni Flash video, Tasks in Gemini (outside the EEA, Switzerland, the UK and Nigeria), Jules (100 tasks a day) and Google Antigravity, Chrome auto browse (US), 1,000 Google Flow credits, Gemini in Gmail, Docs and Vids, 5 TB storage, YouTube Premium Lite, Google Home Premium (Standard), Google Health Premium and $10 monthly Google Cloud credits.

Price
$19.99 per month (US) 21.99 € per month (DE) 219.99 € per year (DE, Google AI Pro (5 TB), annual billing)
Models
Gemini 4 Argon full Plans page: "Access to Gemini 4 Argon"; release notes (2026-10-13): "rolling out globally for Google AI Pro & Ultra subscribers in the Gemini app". The help center model page lists Argon as a chat and task model for Google AI Pro and Ultra. Gemini 3.8 Flash full Plans page: "Access to our Flash model" (version no longer named; it named 3.8 Flash on 2026-10-13, see open_questions). Gemini 3.1 Pro full Plans page: "Access to our Pro model"; the FAQ names "our most capable models, like 3.1 Pro" and a 1M token context window. Gemini 3.5 Flash-Lite full Included from the free plan; chats with a Google AI plan continue with Flash-Lite once the usage limit is reached. Nano Banana 2.1 full "Google AI Pro, Plus, Ultra users have access to our best image model, Nano Banana 2.1" Nano Banana Pro limited Only for regenerating images in the Gemini app via "Redo with Pro". Gemini Omni 1.1 Flash (video) full Video generation model "Omni Flash" in the Gemini app and Google Flow; 10-second videos, 18+.
Limits
Gemini app usage: 4x higher than standard limitsGemini app usage (Gemini plans page): Get four times higher usage limits than Free and access to more features like video generation.Access to Pro models (Google One comparison): IncludedLimit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.When the limit is reached: If you have a Google AI plan and reach your usage limit, your chat will continue with Flash-Lite.Context window: 1 million tokensFile uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Audio and video upload length: Video: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 1 hour." Audio: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 3 hours."Deep Research: Google AI Pro and Google AI Ultra users get higher limits for how many research reports they can create.Chrome auto browse: With a Google AI Pro plan: up to 20 requests per dayTasks in Gemini (formerly Gemini Spark): You can have up to 15 tasks running at a given time.Jules: Daily Tasks (rolling 24 hours): 100; Concurrent Tasks: 15Jules (Google One help): Jules in AI Pro has higher task limits, higher concurrency limits, and higher access to some of the latest models.Google Antigravity: Entry rate limits to agent model in Google Antigravity, our agentic development platformGoogle Antigravity agent requests (Google One comparison): ExpandedGoogle Flow credits: 1,000/moGoogle Flow Music credits: 10,000/moGoogle Cloud credits: $10 monthly creditsAI credits: If you reach your plan’s limit, Google AI Pro and Google AI Ultra members can purchase AI credits to get extra usage in Google Flow and Google Antigravity.Cloud storage: 5 TB of cloud storage across Gmail, Drive, and Photos
Usage score
7/10 · 4x the standard Gemini limits, 20 Chrome auto browse requests a day, 1,000 Flow credits a month and expanded Jules limits.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#2 ChatGPT Plus OpenAI

Mid-tier plan for advanced work: adds GPT-6 reasoning levels in Chat, legacy models, expanded uploads, deep research and memory, and full Codex and ChatGPT Work access (including limited GPT-6 Astra usage and GPT-6.1 Terra) with five-hour and weekly usage windows.

Price
$20.00 per month (US, Billed monthly; no annual billing) 23.00 € per month (DE, Shown as '23 €' / month on the German pricing page)
Models
GPT-6 Sol full Powers GPT-6 in Chat (Instant, Medium and High; Extra High and Pro levels are not included on Plus) and is available in Work and Codex. Pricing page: 'Yes'. GPT-6.1 Sol full ChatGPT Work and Codex only, not Chat. Pricing page: 'Yes'. GPT-6.1 Terra full ChatGPT Work and Codex only, not Chat; new since Oct 26, 2026: 'available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex'. Not yet in the pricing page comparison table or the model catalog. GPT-6 Astra limited Included in ChatGPT Work and Codex only; Plus includes limited Astra usage with optional credits afterward and does not include GPT-6 Pro in Chat. The English pricing page comparison shows 'Yes' for Plus. GPT-6 Luna full ChatGPT Work and Codex. Pricing page: 'Yes'. GPT-5.6 Sol full Selectable in Chat; also in Work and Codex. Pricing page: 'Yes'. GPT-5.6 Terra full Work and Codex. Pricing page: 'Yes'. GPT-5.6 Luna full Work and Codex. Pricing page: 'Yes'. GPT-5 Thinking Mini full Pricing page comparison: 'Expanded'. Legacy models full Pricing page comparison: 'Yes'.
Limits
Overall usage: Limits apply (no absolute numbers published for Chat)Everyday text chats: Unlimited* (*Usage must be reasonable and comply with our policies)Work and Codex, GPT-6 Astra (local messages per 5 hours): 5-45 (estimate, not a fixed message limit; weekly limits may also apply)Work and Codex, GPT-6.1 Sol (local messages per 5 hours): 15-160 (estimate, not a fixed message limit; weekly limits may also apply)Work and Codex, GPT-6.1 Terra (local messages per 5 hours): 35-285 (estimate, not a fixed message limit; weekly limits may also apply)Work and Codex, GPT-6 Sol (local messages per 5 hours): 15-150 (estimate, not a fixed message limit; weekly limits may also apply)Work and Codex, GPT-6 Luna (local messages per 5 hours): 350-3,000 (estimate, not a fixed message limit; weekly limits may also apply)Voice in Chat (Live, rolling 24-hour period): 3 hours with GPT-Live-1.1Deep research: Expanded deep research; usage varies by plan (no number published)GPT Instant total context window: 54KGPT Instant input maximum: ~40 pages of textGPT Reasoning total context window: 256KGPT Reasoning input maximum: ~320 pages of text
Usage score
7/10 · Unlimited text chats within reasonable use, published estimates of 15-160 GPT-6.1 Sol messages per 5 hours in Work and Codex, 3 hours of voice a day.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#3 Claude Pro Anthropic

Entry paid plan for regular users: at least 5x the Free plan's usage per 5-hour session plus a weekly limit, adding Opus, Claude Code, Research, Claude in Chrome and Cowork; Fable models only via paid usage credits.

Price
$20.00 per month (US, "$20 if billed monthly"; prices shown don't include applicable tax) $200.00 per year (US, $17 per month with annual subscription discount ("$200 billed up front"); excl. applicable tax)
Models
Claude Opus 5.5 full Pricing table: Opus "Yes" for Pro; Opus 5.5 is listed for chat on paid plans (1M-token context). Claude Sonnet 5.5 full Pricing table: Sonnet "Yes" for Pro; Sonnet 5.5 is listed for chat on paid plans (1M-token context). Claude Haiku 5.5 full Pricing table: Haiku "Yes" for Pro; Haiku 5.5 is listed for chat on paid plans (1M-token context). Claude Fable 5.1 limited "Fable 5 and Fable 5.1 aren't included in your plan's usage limits"; usable only with usage credits, billed at standard API rates.
Limits
Usage compared with Free: "at least 5x more usage per 5-hour session than Free"; comparison table: "Limit every 5 hours" 1x, "Limit every week" 1x (the reference for Max and Ultra; no absolute number published)Session limit: "a session limit that resets every five hours"; "The period starts when you send your first message, and your limit resets in full five hours later."Weekly limit: "Pro plans have a five-hour limit and a weekly usage limit that applies across all models. Weekly limits reset at a fixed time each week that is assigned to your account."Message count: "there's no fixed message count"; it depends on conversation length and complexity, the model and the features usedClaude Code: "These limits are shared across Claude and Claude Code, meaning all activity in both tools counts against the same limits."Fable models: "Fable 5 and Fable 5.1 aren't included in your plan's usage limits" (available with usage credits)Beyond the limits: Optional usage credits: "Usage credits are billed at standard API rates."
Usage score
6/10 · At least 5x the Free usage per 5-hour session plus a weekly cap, shared with Claude Code; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#4 Perplexity Pro Perplexity

Paid tier for regular research: a model picker covering OpenAI, Anthropic, Google, xAI and other models, more Pro Searches and Research queries, image generation and limited video generation, Projects, and Perplexity Computer with a weekly included amount of usage (Light effort without credits) plus a one-time credit bonus.

Price
$20.00 per month (US, "Perplexity Pro: $20/month or $200/year"; the hub FAQ states the same.) $200.00 per year (US, Billed annually; the pricing page shows "$17/month when billed annually".)
Models
Best (automatic model selection) full “Best” mode is available without quota limits. GPT-6.1 Sol full OpenAI; Search model selector (help-center table as of October 27, 2026), subject to weekly Pro Search limits. Gemini 3.8 Flash full Google; Search model selector. Claude Sonnet 5.5 full Anthropic; Search model selector. Kimi K3 full Moonshot AI; Search model selector. GLM 5.3 full Z.ai; Search model selector. Grok 4.7 full xAI; Search model selector. Nemotron 3.5 Super VL full NVIDIA; Search model selector. GPT-6 Astra limited OpenAI; Perplexity Computer only (not in the Pro Search model selector). 'Eligible Pro and Max subscribers can use GPT-6 Astra for Computer tasks'; availability depends on the plan and data-retention eligibility. Gemini 4 Argon limited Google; Perplexity Computer only (in Search it is Max only). Claude Fable 5.1 limited Anthropic; Perplexity Computer only (not in the Pro Search model selector): 'Pro and Max subscribers can choose Claude Fable 5.1 for demanding Computer tasks.' GPT Image 1 limited Image generation model (OpenAI); limited amount of high-quality images plus medium-quality options. Nano Banana limited Image generation model (Google); Nano Banana Pro is reserved for Max and Enterprise Max. Seedream 4.5 limited Image generation model (ByteDance). Nano Banana 2.1 limited Image generation and editing model (Google) in Perplexity Computer only; 'available to everyone with Computer access' (changelog, October 19, 2026). Veo 3.1 limited Video generation (Google), assigned automatically; videos up to 8 seconds; smaller monthly allocation than Max. MiniMax H3 limited Video generation model (MiniMax) in Perplexity Computer only; 'Pro and Max subscribers on personal accounts can produce finished videos' (changelog, October 5, 2026). Seedance 2.5 limited Video generation model (ByteDance) in Perplexity Computer only, for Pro and Max subscribers on personal accounts (changelog, October 5, 2026).
Limits
Pro Searches: Weekly limits (average use)Pro Search use: Pro Search is no longer a separate toggle, so there’s nothing to turn on. [...] Perplexity uses Pro Search automatically when it’s included in your plan, up to your plan’s limit.Research Queries: Monthly limits (average use)Research limit reset: Research limits reset on a rolling basis rather than on a fixed calendar date. [...] access returns automatically as older Research queries roll out of the usage window.Browser Agent (Comet Assistant) Queries: Monthly limits (average use)File and App Creation: Monthly limits (average use); "Limited amount Create files and apps queries every 30 days"File Uploads (sessions): Weekly limits (average use)File upload allowance: File uploads are not unlimited on Free, Pro, or Max plans. Each plan includes an allowance of file uploads that resets on a rolling weekly basis, and Max includes a higher allowance than Pro. Images you attach or paste into a search count toward the same allowance. [...] There is no fixed weekly reset day.File uploads per project: Up to 50 file uploads per projectFile size: File size is limited to 40 MB for all file types.Best mode: “Best” mode is available without quota limits.Advanced AI model queries: For Advanced AI model queries, access may be limited during weeks of especially heavy usage.Computer monthly credits: No monthly allocation (paid plans include weekly Computer usage)Computer weekly included usage: Paid Consumer Pro plans now include a weekly amount of Computer usage. [...] Your included usage resets every week. [...] Unused included usage does not roll over from one week to the next. (amount not published; "Included usage is rolling out gradually"; "It is not included with promotional or non-paying Pro plans")Computer effort without credits: Included usage but no credits: Computer runs your tasks on the Light effort setting. Other effort settings are unavailable until you have credits.Computer new-subscriber bonus: 4,000 credits (one-time, expire in about 30 days)Credit spending cap: Pro accounts also have a monthly spending cap. (amount not published)Image generation: Pro/Enterprise Pro: Unlock access to limited amount of high-quality images, plus additional medium-quality options.Video generation: We are unable to provide exact numbers ... Max subscribers receive the highest number of video generation credits each month, while Pro users are allocated a smaller amount. (credits replenish on a rolling 30-day basis)
Usage score
5/10 · Weekly and monthly limits described only as 'average use', with small allowances for image and video generation.
  • Web search
  • Deep research
  • Image generation
  • Video generation ("Limited access to video generation tools": Veo 3.1, videos up to 8 seconds, monthly allocation not published and smaller than Max. In Perplexity Computer, Pro subscribers on personal accounts can also create videos with MiniMax H3 and Seedance 2.5.)
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#5 Microsoft 365 Premium Microsoft

Microsoft 365 Family plus extensive Copilot usage and the Researcher and Analyst agents exclusive to Premium and Pro, with limited usage of the Cowork agent (preview, US and Canada); it replaces Copilot Pro, which is no longer sold. AI features remain limited to the subscription owner.

Price
$19.99 per month (US, '$19.99 /month'; 'Try free for 1 month'; renews automatically unless canceled.) $199.99 per year (US, '$199.99 /year') 22.00 € per month (DE, Shown as '22,00 € /Monat'; '1 Monat lang kostenlos testen'. VAT is not mentioned on the page.) 219.00 € per year (DE, Shown as '219,00 € /Jahr'.)
Models
GPT-5 full Smart mode: 'Smart mode is powered by GPT-5'. GPT-5 is not in the OpenAI catalog. Models used by Researcher, Analyst and Cowork are not named for personal accounts. Think Deeper (OpenAI reasoning models, not named) full 'Think Deeper is powered by the latest reasoning models from OpenAI.' Premium users 'have priority access whenever capacity is limited'. MAI-Image-2 full Microsoft AI, Mar 19, 2026: 'MAI-Image-2 is beginning to roll out on Copilot and Bing Image Creator.' Image generation: 'Extensive usage' on Premium. MAI-Image-2-Efficient full Microsoft AI, Apr 14, 2026: 'It's also rolling out across Copilot and Bing'. The model serving a given request is not stated.
Limits
Chat: Extensive usageDeep reasoning: Extensive usagePages: Extensive usageAudio Overview: Extensive usageVoice: More usageVision: More usageImage gen: Extensive usageResearcher: Extensive usageAnalyst: Extensive usageCowork (in preview): Limited usage; Currently only available in the US and Canada. Availability varies by platform and device.Priority access: YesAI credits: Microsoft 365 Personal, Family, Premium, and Pro plans now include AI credits (no allotment published)Agentic AI (pricing page): Limited usage for agentic AI that gets things done for you including advanced research and analyticsOverall usage (pricing page): Advanced Copilot chat with extensive usage; Extensive usage for visual AI assistance from your screen or camera; Extensive usage for AI-generated audio (English only); Extensive usage for AI image creation in Copilot; Access to Copilot features currently exclusive to Premium & ProAI usage limits (FAQ): Microsoft 365 Premium includes higher limits (than Personal and Family); no numbers publishedWhen a limit is reached: If you reach a feature limit or use all of your AI credits, that feature may be unavailable until your allowance resets.Users, devices, storage: Everything in Microsoft 365 Family: For 1 to 6 people (AI for subscription owner only); up to 5 devices per person; 2 TB of shared, secure cloud storageAI features: AI features only available to subscription owner and cannot be shared; usage limits apply.
Usage score
6/10 · Extensive usage for chat, reasoning, images, Researcher and Analyst in Microsoft's table; the pricing page calls agentic AI usage limited.
  • Web search
  • Deep research (Researcher agent ('Extensive usage' in the support limits table); the US pricing page says 'Limited usage for agentic AI ... including advanced research and analytics' (Pro: 'Highest usage'); in the support table only Cowork is 'Limited usage' for Premium.)
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent (Cowork (in preview): 'Limited usage' in the support limits table (Pro: 'Highest usage'), currently only in the US and Canada; Browse with Copilot in Edge is rolling out for Premium in the US. Pricing page: 'Limited usage for agentic AI'.)
  • Custom assistants or projects
  • Memory across chats

#6 The X Subscription Premium xAI

Mid tier of The X Subscription and xAI's recommendation for new AI users. Over Plus it adds 4x the usage, the latest Grok models, faster Chat responses, image and video generation, Grok Bot, Cursor cloud agents, Bugbot code reviews and Origin code hosting, and full X Premium+ benefits once a personal X account is connected.

Price
$30.00 per month (US, excl. taxes ('Prices shown do not include applicable taxes or fees (e.g. VAT)')) $300.00 per year (US, billed annually ($25/mo); list price excl. sales tax) 31.00 € per month (DE, regular price; page showed an intro offer '50% Rabatt für 1 Monat, danach 31 € monatlich' (16 € shown for the first month on 2026-11-01). Signed-out visitor in Germany; prices shown excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren')) 310.00 € per year (DE, 310 €/Jahr (26 €/Monat), as shown on grok.com/subscription in Germany; excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren'))
Models
Grok 4.7 full x.ai/pricing lists 'Latest Grok models' from Premium up; grok.com release notes (Oct 3, 2026): Grok 4.7 'now powers Expert mode on grok.com, and Fast mode for paid subscribers'. Grok Imagine (images and video) full Image and video generation included ('Make stunning AI images & videos'). The Imagine model per plan is not stated.
Limits
Included usage: 4x more usage compared to PlusSpeed: Faster Chat responses ('Schnellere Chat-Antworten' on grok.com/subscription)Usage pool: Every tier includes a pool of AI usage shared across Grok and Cursor. Within Grok, the pool covers Chat, Imagine, Voice, and Bot. Higher tiers include larger pools, and usage costs vary by product.Grok API: Paid plans include Grok API access. API usage is billed separately at standard model rates and doesn't draw from your plan's usage pool.Refresh: Your allowance refreshes weekly, so one heavy day never wipes out your whole month.Grok Bot: Bot usage draws from that seat’s shared weekly usage pool, not a separate additive pool.Grok Bot usage: Weekly Grok Bot usage includedVideo resolution: 720p videos automatically fall back to 480p once you hit the 720p cap for your tier.Beyond included usage: If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh.Usage limits: not published (no absolute numbers)
Usage score
6/10 · 4x the Plus usage pool, shared across Grok and Cursor and refreshing weekly; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

Advanced plans (about 100 a month)

# Plan Score Price (US) Price (DE) Best included model (Text Arena) 40 %Features 20 %Usage allowance (editorial, 0–10) 20 %
#1 Google AI Ultra (5x)Google 94.0 $99.99 99.99 € Gemini 4 Argon (1,517)100/100 10 of 10100/100 7/1070/100
#2 Claude MaxAnthropic 90.5 $100.00 – Claude Opus 5.5 (1,507)96/100 8 of 1080/100 8/1080/100
#3 ChatGPT Pro 100OpenAI 88.6 $100.00 103.00 € GPT-6.1 Sol (1,480)87/100 9 of 1090/100 8/1080/100
#4 Microsoft 365 ProMicrosoft 83.3 $99.99 110.00 € No data 8 of 1080/100 7/1070/100
#5 The X Subscription SuperxAI 81.0 $100.00 105.00 € Grok 4.7 (1,441)73/100 9 of 9100/100 6/1060/100

#1 Google AI Ultra (5x) Google

Entry Ultra tier with 5x the Gemini usage limits of Pro: the highest access to Gemini 4 Argon, Deep Think (being replaced by Gemini 4 Argon), first access to Tasks in Gemini (select countries), Jules (300 tasks a day) and higher Antigravity limits, Chrome auto browse (US), 10,000 Google Flow credits, Project Genie, 20 TB storage, YouTube Premium (individual), Google Home Premium (Advanced) and $40 monthly Google Cloud credits.

Price
$99.99 per month (US, 5x usage tier) 99.99 € per month (DE, 5x usage tier)
Models
Gemini 4 Argon full Plans page: "Highest access to Gemini 4 Argon"; release notes (2026-10-13): rolling out globally for Google AI Pro & Ultra subscribers in the Gemini app. Google One card: "Higher access to Gemini 4 Argon". Gemini 3.8 Flash full Plans page: "Highest access to our Flash model" (version no longer named; it named 3.8 Flash on 2026-10-13, see open_questions). Gemini 3.1 Pro full Plans page: "Highest access to ... Pro model"; the Ultra FAQ names 3.1 Pro. Google One comparison: "Higher" access to Pro models; 1M-token context window. Gemini Deep Think full Help center: available to Google AI Ultra subscribers (18+); select the Pro model, then Thinking Level > Deep Think. "Deep Think is being replaced by Gemini 4 Argon." Gemini 3.5 Flash-Lite full Included from the free plan; chats with a Google AI plan continue with Flash-Lite once the usage limit is reached. Nano Banana 2.1 full "Google AI Pro, Plus, Ultra users have access to our best image model, Nano Banana 2.1" Nano Banana Pro limited Only for regenerating images in the Gemini app via "Redo with Pro". Gemini Omni 1.1 Flash (video) full Video generation model "Omni Flash" with higher limits in the Gemini app and Google Flow; 10-second videos, 18+.
Limits
Gemini app usage: $99.99/ month: 5x higher usage limits vs. AI ProGemini app usage (help center): 5x or 20x higher than AI Pro limits depending on your subscriptionGemini and Google Antigravity usage (Google One help): Based on your specific Google AI Ultra plan, you get 5x or 20x usage quota in Gemini and Google Antigravity compared to the Google AI Pro plan.Access to Pro models (Google One comparison): HigherLimit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.When the limit is reached: If you have a Google AI plan and reach your usage limit, your chat will continue with Flash-Lite.Context window: 1 million tokensDeep Think context: If you use Deep Think, there is a 192 thousand token window.File uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Audio and video upload length: Video: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 1 hour." Audio: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 3 hours."Deep Research: Google AI Pro and Google AI Ultra users get higher limits for how many research reports they can create.Chrome auto browse: With a Google AI Ultra plan: up to 200 requests per dayTasks in Gemini (formerly Gemini Spark): You can have up to 15 tasks running at a given time.Jules: Daily Tasks (rolling 24 hours): 300; Concurrent Tasks: 60 ("Jules in Ultra"; the docs do not distinguish the 5x and 20x tiers)Jules (Google One help): Jules with the highest task limits, highest concurrency limits, and priority model accessGoogle Antigravity: Ultra 5x: Higher rate limits to agent model in Google Antigravity, our agentic development platformGoogle Antigravity agent requests (Google One comparison): HigherGoogle Flow credits: 10,000/moGoogle Flow Music credits: 30,000/moGoogle Cloud credits: $40 monthly creditsAI credits: If you reach your plan’s limit, Google AI Pro and Google AI Ultra members can purchase AI credits to get extra usage in Google Flow and Google Antigravity.Cloud storage: Google AI Ultra 5x (20 TB of storage)
Usage score
7/10 · 5x the AI Pro limits, 200 Chrome auto browse requests a day, 10,000 Flow credits and $40 of Google Cloud credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#2 Claude Max Anthropic

For people who use Claude throughout the day: five times Pro's usage every five hours and every week, Fable models included up to 50% of the weekly limit, priority access at high-traffic times and $100 a month in API credits.

Price
$100.00 per month (US, Web price; monthly subscription only; mobile pricing may vary; excl. applicable tax)
Models
Claude Opus 5.5 full Pricing table: Opus "Yes" for Max; Opus 5.5 is listed for chat on paid plans (1M-token context). Claude Sonnet 5.5 full Pricing table: Sonnet "Yes" for Max; Sonnet 5.5 is listed for chat on paid plans (1M-token context). Claude Haiku 5.5 full Pricing table: Haiku "Yes" for Max; Haiku 5.5 is listed for chat on paid plans (1M-token context). Claude Fable 5.1 full Included as a standard part of the plan, capped at "up to 50% of your weekly usage limits"; beyond that via usage credits.
Limits
Usage compared with Pro: "Get five times the Pro plan's usage every five hours and every week." (pricing table: 5x every 5 hours, 5x every week; no absolute number published)Session limit: "a session limit that resets every five hours"; "The period starts when you send your first message, and your limit resets in full five hours later."Weekly limit: "Max plans have a five-hour limit, a weekly usage limit that applies across all models, and a separate weekly limit for Fable. The weekly limit resets at a fixed time each week that is assigned to your account."Fable models: "You can use up to 50% of your weekly usage limits on Fable models at no extra cost."Claude Code: "These limits are shared across Claude and Claude Code, meaning all activity in both tools counts against the same limits."Monthly API credits: $100/month ("New Monthly API credits" for building on the Claude Platform)Beyond the limits: Optional usage credits: "Usage credits are billed at standard API rates."
Usage score
8/10 · Five times the Pro usage every five hours and every week, a separate weekly limit for Fable (up to half of it) and $100 of API credits a month; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#3 ChatGPT Pro 100 OpenAI

Entry Pro tier for research and coding: everything in Plus plus GPT-6 Pro reasoning powered by GPT-6 Astra, Extra High thinking, larger context, longer Codex and Work sessions without a five-hour limit (allowance doubled on October 29, 2026), maximum deep research, Dot and $100 a month in API credits.

Price
$100.00 per month (US, Billed monthly; annual billing and multi-month prepayment are not available) 103.00 € per month (DE, Shown as 'Ab 103 €' / month ('From') on the German pricing page Pro card; Pro 100 is the lowest Pro tier)
Models
GPT-6 Astra full Powers GPT-6 Pro in Chat (Pro $100 is listed as eligible) and is included in Work and Codex; GPT-6 Pro has its own usage limits. Pricing page: 'Expanded'. GPT-6 Sol full Powers GPT-6 Instant through Extra High in Chat; also in Work and Codex. Pricing page: 'Expanded'. GPT-6.1 Sol full ChatGPT Work and Codex only, not Chat. Pricing page: 'Expanded'. GPT-6.1 Terra full ChatGPT Work and Codex only, not Chat; new since Oct 26, 2026: 'available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex'. Not yet in the pricing page comparison table or the model catalog. GPT-6 Luna full Work and Codex. Pricing page: 'Expanded'. GPT-5.6 Sol full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Sol Pro full Pricing page: 'Yes'. On Pro $100, GPT-6 Pro and GPT-5.6 Sol Pro draw from one weekly allowance. GPT-5.6 Terra full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Luna full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5 Thinking Mini full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' Legacy models full Pricing page comparison: 'Yes'.
Limits
Overall usage (Chat): Unlimited subject to abuse guardrailsIncluded usage vs. other Pro tiers: Pro 200 includes more usage than Pro 100. Pro 500 offers the highest included usage of the three plans.GPT-6 Pro and GPT-5.6 Sol Pro (Chat): Both models draw from one weekly allowance (no number published); a Pro subscription does not include unlimited use of the GPT-6 Pro modelVoice in Chat (Live, rolling 24-hour period): 15 hours with GPT-Live-1.1Work and Codex: No five-hour usage limit in Work and Codex; the plan still has an included usage allowance (no number published)Work and Codex allowance change (Oct 29, 2026): 'Pro 100 and Pro 500 now include twice their previous Work and Codex allowance.' (no absolute number published)Deep research: Maximum deep research; usage varies by plan (no number published)GPT Instant total context window: 128KGPT Instant input maximum: ~250 pages of textGPT Reasoning total context window: 400KGPT Reasoning input maximum: ~680 pages of textMonthly API credits: $100 per monthly billing period (promotional OpenAI API credits for one linked API organization; they 'do not increase your ChatGPT or signed-in Codex usage limits')
Usage score
8/10 · Unlimited chat subject to abuse guardrails, no five-hour limit in Work and Codex and 15 hours of voice a day; GPT-6 Pro has a weekly cap.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#4 Microsoft 365 Pro Microsoft

Top individual tier with monthly billing only: everything in Microsoft 365 Premium plus the highest Copilot usage limits (chat, voice, vision and agentic AI, including 'Highest usage' for the Cowork preview where Premium has 'Limited usage') and 5 TB of shared storage. AI features are for the subscription owner only.

Price
$99.99 per month (US, '$99.99 /month'; 'Monthly billing only.' No annual price and no free trial shown (store page also shows '$99.99/month').) 110.00 € per month (DE, Shown as '110,00 € /Monat'; 'Nur Monatliche Abrechnung.' VAT is not mentioned on the page.)
Models
GPT-5 full Smart mode: 'Smart mode is powered by GPT-5'. GPT-5 is not in the OpenAI catalog. Models used by Researcher, Analyst and Cowork are not named for personal accounts. Think Deeper (OpenAI reasoning models, not named) full 'Think Deeper is powered by the latest reasoning models from OpenAI.' 'Use of Think Deeper in Copilot is not limited'. MAI-Image-2 full Microsoft AI, Mar 19, 2026: 'MAI-Image-2 is beginning to roll out on Copilot and Bing Image Creator.' Image generation: 'Extensive usage' on Pro. MAI-Image-2-Efficient full Microsoft AI, Apr 14, 2026: 'It's also rolling out across Copilot and Bing'. The model serving a given request is not stated.
Limits
Chat: Extensive usageDeep reasoning: Extensive usagePages: Extensive usageAudio Overview: Extensive usageVoice: Highest usageVision: Highest usageImage gen: Extensive usageResearcher: Extensive usageAnalyst: Extensive usageCowork (in preview): Highest usage; Currently only available in the US and Canada. Availability varies by platform and device.Priority access: YesAI credits: Microsoft 365 Personal, Family, Premium, and Pro plans now include AI credits (no allotment published)Agentic AI (pricing page): Highest usage for agentic AI that gets things done for you including advanced research and analyticsOverall usage (pricing page): Advanced Copilot chat with highest usage; Highest usage for visual AI assistance from your screen or camera; Highest usage for AI-generated audio (English only); comparison table: Overall usage amounts 'Highest', AI image generation 'Highest'AI usage limits (FAQ): Microsoft 365 Pro offers the highest limits; no numbers publishedUpgrade path (support page): If you already have a Microsoft 365 Personal, Family, or Premium plan, you can switch to Microsoft 365 Pro to access the highest usage limits available.Store page: Get the highest usage for Copilot, including agentic AI features; Advanced Copilot chat with highest usage; Extensive usage to create AI imagesWhen a limit is reached: If you reach a feature limit or use all of your AI credits, that feature may be unavailable until your allowance resets.Users, devices, storage: Everything in Microsoft 365 Premium, plus 5 TB of shared, secure cloud storage; up to 6 people (AI for subscription owner only)AI features: AI features only available to subscription owner and cannot be shared; usage limits apply.
Usage score
7/10 · The plan Microsoft describes as having "the highest usage limits available": highest usage for Copilot including Cowork, extensive image creation and AI credits; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#5 The X Subscription Super xAI

High-usage tier of The X Subscription, recommended by xAI for daily agent users. It includes everything in Premium with 15x the usage of Plus ('Significantly higher usage across Chat, Imagine, Voice & Build'), priority access at peak times, 1080p videos up to 15 seconds and 'Lightning-fast replies'.

Price
$100.00 per month (US, excl. taxes ('Prices shown do not include applicable taxes or fees (e.g. VAT)')) $1000.00 per year (US, billed annually ($84/mo); list price excl. sales tax) 105.00 € per month (DE, regular price; page showed an intro offer '50% Rabatt für 1 Monat, danach 105 € monatlich' (53 € shown for the first month on 2026-11-01). Signed-out visitor in Germany; prices shown excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren')) 1050.00 € per year (DE, 1.050 €/Jahr (88 €/Monat), as shown on grok.com/subscription in Germany; excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren'))
Models
Grok 4.7 full x.ai/pricing lists 'Latest Grok models' from Premium up; grok.com release notes (Oct 3, 2026): Grok 4.7 'now powers Expert mode on grok.com, and Fast mode for paid subscribers'. Grok Imagine (images and video) full Image and video generation included. The Imagine model per plan is not stated.
Limits
Included usage: 15x more usage compared to PlusUsage across products: Significantly higher usage across Chat, Imagine, Voice & BuildPeak times: Priority access at peak times (grok.com/subscription: 'Bevorzugter Modell-Zugriff bei hoher Auslastung')Speed: Lightning-fast replies (grok.com/subscription: 'Schnellere Antworten & Generierungen')Usage pool: Every tier includes a pool of AI usage shared across Grok and Cursor. Within Grok, the pool covers Chat, Imagine, Voice, and Bot. Higher tiers include larger pools, and usage costs vary by product.Grok API: Paid plans include Grok API access. API usage is billed separately at standard model rates and doesn't draw from your plan's usage pool.Refresh: Your allowance refreshes weekly, so one heavy day never wipes out your whole month.Grok Bot: Bot usage draws from that seat’s shared weekly usage pool, not a separate additive pool.Grok Bot usage: 3.5x Premium usageVideo resolution: 720p videos automatically fall back to 480p once you hit the 720p cap for your tier.1080p video: Create 1080p, 15s videos (grok.com/subscription: 'Erstelle 1080p Videos'); listed from Super upBeyond included usage: If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh.Usage limits: not published (no absolute numbers)
Usage score
6/10 · 15x the Plus usage pool with priority access at peak times, refreshing weekly; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

Premium plans (200 a month and more)

# Plan Score Price (US) Price (DE) Best included model (Text Arena) 40 %Features 20 %Usage allowance (editorial, 0–10) 20 %
#1 Google AI Ultra (20x)Google 96.0 $199.99 219.99 € Gemini 4 Argon (1,517)100/100 10 of 10100/100 8/1080/100
#2 Claude UltraAnthropic 92.5 $200.00 – Claude Opus 5.5 (1,507)96/100 8 of 1080/100 9/1090/100
#3 Perplexity MaxPerplexity 92.0 $200.00 – Gemini 4 Argon (1,517)100/100 9 of 1090/100 7/1070/100
#4 ChatGPT Pro 200OpenAI 90.6 $200.00 – GPT-6.1 Sol (1,480)87/100 9 of 1090/100 9/1090/100
#5 The X Subscription UltraxAI 85.0 $200.00 210.00 € Grok 4.7 (1,441)73/100 9 of 9100/100 8/1080/100
#6 ChatGPT Pro 500OpenAI 78.6 $500.00 – GPT-6.1 Sol (1,480)87/100 9 of 1090/100 9/1090/100

#1 Google AI Ultra (20x) Google

Top Ultra tier with 20x the Gemini usage limits of Pro: everything in Ultra (5x) with the highest Jules and Antigravity limits, 25,000 Google Flow credits, 30 TB storage and $100 monthly Google Cloud credits, plus the highest access to Gemini 4 Argon, Deep Think (being replaced by Gemini 4 Argon), first access to Tasks in Gemini (select countries), Project Genie, YouTube Premium (individual) and Google Home Premium (Advanced).

Price
$199.99 per month (US, 20x usage tier) 219.99 € per month (DE, 20x usage tier)
Models
Gemini 4 Argon full Plans page: "Highest access to Gemini 4 Argon"; release notes (2026-10-13): rolling out globally for Google AI Pro & Ultra subscribers in the Gemini app. Google One card: "Higher access to Gemini 4 Argon". Gemini 3.8 Flash full Plans page: "Highest access to our Flash model" (version no longer named; it named 3.8 Flash on 2026-10-13, see open_questions). Gemini 3.1 Pro full Plans page: "Highest access to ... Pro model"; the Ultra FAQ names 3.1 Pro. Google One comparison: "Highest" access to Pro models; 1M-token context window. Gemini Deep Think full Help center: available to Google AI Ultra subscribers (18+); select the Pro model, then Thinking Level > Deep Think. "Deep Think is being replaced by Gemini 4 Argon." Gemini 3.5 Flash-Lite full Included from the free plan; chats with a Google AI plan continue with Flash-Lite once the usage limit is reached. Nano Banana 2.1 full "Google AI Pro, Plus, Ultra users have access to our best image model, Nano Banana 2.1" Nano Banana Pro limited Only for regenerating images in the Gemini app via "Redo with Pro". Gemini Omni 1.1 Flash (video) full Video generation model "Omni Flash" with higher limits in the Gemini app and Google Flow; 10-second videos, 18+.
Limits
Gemini app usage: $199.99 / month: 20x higher usage limits vs. AI ProGemini app usage (help center): 5x or 20x higher than AI Pro limits depending on your subscriptionGemini and Google Antigravity usage (Google One help): Based on your specific Google AI Ultra plan, you get 5x or 20x usage quota in Gemini and Google Antigravity compared to the Google AI Pro plan.Access to Pro models (Google One comparison): HighestLimit refresh: Your limit refreshes every 5 hours until you reach your weekly limit.When the limit is reached: If you have a Google AI plan and reach your usage limit, your chat will continue with Flash-Lite.Context window: 1 million tokensDeep Think context: If you use Deep Think, there is a 192 thousand token window.File uploads: Up to 10 files (subject to availability) can be uploaded in the same prompt. Each video can be up to 2 GB. All other supported file types can be up to 100 MB.Audio and video upload length: Video: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 1 hour." Audio: "Upgrade to Google AI Pro or Google AI Ultra to extend total upload length to 3 hours."Deep Research: Google AI Pro and Google AI Ultra users get higher limits for how many research reports they can create.Chrome auto browse: With a Google AI Ultra plan: up to 200 requests per dayTasks in Gemini (formerly Gemini Spark): You can have up to 15 tasks running at a given time.Jules: Daily Tasks (rolling 24 hours): 300; Concurrent Tasks: 60 ("Jules in Ultra"; the docs do not distinguish the 5x and 20x tiers)Jules (Google One help): Jules with the highest task limits, highest concurrency limits, and priority model accessGoogle Antigravity: Ultra 20x: Highest rate limits to agent model in Google Antigravity, our agentic development platformGoogle Antigravity agent requests (Google One comparison): HighestGoogle Flow credits: 25,000/moGoogle Flow Music credits: 30,000/moGoogle Cloud credits: $100 monthly creditsAI credits: If you reach your plan’s limit, Google AI Pro and Google AI Ultra members can purchase AI credits to get extra usage in Google Flow and Google Antigravity.Cloud storage: Google AI Ultra 20x (30 TB of storage)
Usage score
8/10 · 20x the AI Pro limits, 200 Chrome auto browse requests a day, 25,000 Flow credits and $100 of Google Cloud credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#2 Claude Ultra Anthropic

Highest personal tier, formerly Max 20x: 10 times Pro's weekly usage with no five-hour limit for verified accounts, everything in Max including Fable up to 50% of the weekly limit, and $200 a month in API credits.

Price
$200.00 per month (US, Web price; monthly subscription only; mobile pricing may vary; excl. applicable tax)
Models
Claude Opus 5.5 full Pricing table: Opus "Yes" for Ultra; Opus 5.5 is listed for chat on paid plans (1M-token context). Claude Sonnet 5.5 full Pricing table: Sonnet "Yes" for Ultra; Sonnet 5.5 is listed for chat on paid plans (1M-token context). Claude Haiku 5.5 full Pricing table: Haiku "Yes" for Ultra; Haiku 5.5 is listed for chat on paid plans (1M-token context). Claude Fable 5.1 full Included as a standard part of the plan, capped at "up to 50% of your weekly usage limits"; beyond that via usage credits.
Limits
Usage compared with Pro: "Get 10 times the Pro plan's usage every week." (pricing table: "Limit every week" 10x; no absolute number published)Five-hour limit: "Verified Ultra accounts don't have a five-hour limit. If your account isn't verified yet, a temporary limit applies." Accounts are verified automatically "using factors like your account and payment history".Weekly limit: "Ultra plans have a weekly usage limit that applies across all models, and a separate weekly limit for Fable."Fable models: "You can use up to 50% of your weekly usage limits on Fable models at no extra cost."Claude Code: "These limits are shared across Claude and Claude Code, meaning all activity in both tools counts against the same limits."Monthly API credits: $200/month ("$200 a month in API credits, up from $100 on Max", for building on the Claude Platform)Beyond the limits: Optional usage credits: "Usage credits are billed at standard API rates."
Usage score
9/10 · Ten times the Pro weekly usage with no five-hour limit for verified accounts, a separate weekly limit for Fable and $200 of API credits a month; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#3 Perplexity Max Perplexity

Top consumer tier for heavy users: adds GPT-6 Astra, Gemini 4 Argon, Claude Opus 5.5 and Claude Fable 5.1 to the Search model selector, Model Council, Max Assistant in Comet, 10,000 Perplexity Computer credits per month, Brain (preview), priority support and the highest limits for research, uploads, images and video.

Price
$200.00 per month (US, "Perplexity Max costs $200/monthly or $2000/annually.") $2000.00 per year (US, Billed annually; annual billing only on the web app. The pricing page shows "$167/month when billed annually".)
Models
Best (automatic model selection) full Automatic model choice inside Search. GPT-6.1 Sol full OpenAI; Search model selector (help-center table as of October 27, 2026). Ultrafast speed in Perplexity Computer on web is Max only (changelog, October 19, 2026). GPT-6 Astra full OpenAI; in the Search model selector Max only among consumer plans. Eligible Pro and Max subscribers can also use it in Perplexity Computer (changelog, September 21, 2026); Ultrafast speed in Computer on web is Max only (changelog, October 19, 2026). Gemini 4 Argon full Google; in the Search model selector Max only among consumer plans; also available in Perplexity Computer for Pro and Max (changelog, October 19, 2026). Gemini 3.8 Flash full Google; Search model selector. Claude Sonnet 5.5 full Anthropic; Search model selector. Claude Opus 5.5 full Anthropic; in the Search model selector Max only among consumer plans. In Perplexity Computer it powers Standard effort for 'eligible paid personal accounts' (changelog, October 5, 2026). Claude Fable 5.1 full Anthropic; in the Search model selector Max only among consumer plans. Pro and Max subscribers can choose it for Computer tasks (changelog, October 5, 2026). Kimi K3 full Moonshot AI; Search model selector. GLM 5.3 full Z.ai; Search model selector. Grok 4.7 full xAI; Search model selector. Nemotron 3.5 Super VL full NVIDIA; Search model selector. GPT Image 1 full Image generation model (OpenAI); "extensive access to high-quality image generation". Nano Banana full Image generation model (Google). Nano Banana Pro full Image generation model (Google); Max and Enterprise Max only. Seedream 4.5 full Image generation model (ByteDance). Nano Banana 2.1 limited Image generation and editing model (Google) in Perplexity Computer only; 'available to everyone with Computer access' (changelog, October 19, 2026). Veo 3.1 full Video generation (Google), assigned automatically; videos up to 8 seconds; highest monthly allocation of the consumer plans. MiniMax H3 limited Video generation model (MiniMax) in Perplexity Computer only; 'Pro and Max subscribers on personal accounts can produce finished videos' (changelog, October 5, 2026). Seedance 2.5 limited Video generation model (ByteDance) in Perplexity Computer only, for Pro and Max subscribers on personal accounts (changelog, October 5, 2026).
Limits
Overall usage: With high usage limits. (no absolute numbers published)Pro Searches: Weekly limits (advanced use)Pro Search use: Pro Search is no longer a separate toggle, so there’s nothing to turn on. [...] Perplexity uses Pro Search automatically when it’s included in your plan, up to your plan’s limit.Research Queries: Monthly limits (advanced use)Research limit reset: Research limits reset on a rolling basis rather than on a fixed calendar date. [...] access returns automatically as older Research queries roll out of the usage window.Browser Agent (Comet Assistant) Queries: Monthly limits (advanced use)Browser agent queries (Max Assistant on Comet): As a Max subscriber, you receive the highest weekly limit on browser agent queries available across all subscription tiers.File and App Creation: Monthly limits (advanced use)File Uploads (sessions): Weekly limits (advanced use)File upload allowance: File uploads are not unlimited on Free, Pro, or Max plans. Each plan includes an allowance of file uploads that resets on a rolling weekly basis, and Max includes a higher allowance than Pro. Images you attach or paste into a search count toward the same allowance. [...] There is no fixed weekly reset day.File size: File size is limited to 40 MB for all file types.Computer monthly credits: 10,000 credits per monthComputer weekly included usage: Not included: "It is not included with promotional or non-paying Pro plans, Max plans, or Enterprise plans. Those plans keep the credit benefits described on this page."Computer new-subscriber bonus: 35,000 credits (one-time, expire in about 30 days)Credit spending cap: Every Max account has a monthly spending cap on total credit charges per billing period. The default is $200, and you can set it anywhere up to $5,000 in your credit settings.Model Council: There is a monthly cap on how many Model Council queries you can run, except for Model Council in Computer, which uses credits. (number not published)Get answers from the web (pricing table): 10x (Pro column shows a checkmark; baseline not stated)File uploads (pricing table): 10x (Pro column shows a checkmark; baseline not stated)Asset generation (pricing table): 10x (Pro column shows a checkmark; baseline not stated)Video generation (pricing table): 5x (Pro column shows a checkmark; baseline not stated)Image generation: Max/Enterprise Max: Experience our most generous offering, with extensive access to high-quality image generation capabilities.Video generation: Max subscribers receive the highest number of video generation credits each month (exact numbers not provided; credits replenish on a rolling 30-day basis)
Usage score
7/10 · Limits for 'advanced use', the highest browser-agent limit of all tiers and 10,000 Computer credits a month.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#4 ChatGPT Pro 200 OpenAI

Middle Pro tier with the same Pro features as Pro 100 but more included usage, a separate weekly GPT-6 Pro allowance, unlimited GPT-Live-1.1 voice and $200 a month in API credits; since October 29, 2026 all Pro 200 subscriptions, including new ones, get the same higher Work and Codex allowance.

Price
$200.00 per month (US, Billed monthly; annual billing and multi-month prepayment are not available)
Models
GPT-6 Astra full Powers GPT-6 Pro in Chat (Pro $200 is listed as eligible) and is included in Work and Codex; GPT-6 Pro has its own usage limits. Pricing page: 'Expanded'. GPT-6 Sol full Powers GPT-6 Instant through Extra High in Chat; also in Work and Codex. Pricing page: 'Expanded'. GPT-6.1 Sol full ChatGPT Work and Codex only, not Chat. Pricing page: 'Expanded'. GPT-6.1 Terra full ChatGPT Work and Codex only, not Chat; new since Oct 26, 2026: 'available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex'. Not yet in the pricing page comparison table or the model catalog. GPT-6 Luna full Work and Codex. Pricing page: 'Expanded'. GPT-5.6 Sol full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Sol Pro full Pricing page: 'Yes'. On Pro $200, available while the GPT-5.6 Sol Pro daily allowance and combined daily allowance have usage remaining. GPT-5.6 Terra full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Luna full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5 Thinking Mini full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' Legacy models full Pricing page comparison: 'Yes'.
Limits
Overall usage (Chat): Unlimited subject to abuse guardrailsIncluded usage vs. other Pro tiers: Pro 200 includes more usage than Pro 100. Pro 500 offers the highest included usage of the three plans.GPT-6 Pro (Chat): Weekly limit (no number published); when reached, ChatGPT automatically switches to GPT-5.6 Thinking at MediumGPT-6 Pro weekly limit change (Oct 29, 2026): 'On Pro 200, the lower weekly limit now matches the higher one. If you already had the higher limit, it stays the same.'GPT-5.6 Sol Pro (Chat): Available only while your GPT-5.6 Sol Pro daily allowance and combined daily allowance have usage remaining (no number published)Voice in Chat (Live, rolling 24-hour period): Unlimited GPT-Live-1.1 (subject to abuse guardrails)Work and Codex: No five-hour usage limit in Work and Codex; the plan still has an included usage allowance (no number published)Work and Codex allowance change (Oct 29, 2026): The announced Pro 200 reduction is canceled: 'All Pro 200 subscriptions now include the same Work and Codex allowance'; the lower allowance 'has doubled to match', and new subscriptions 'include the increased Work and Codex allowance' (no absolute number published)Deep research: Maximum deep research; usage varies by plan (no number published)GPT Instant total context window: 128KGPT Instant input maximum: ~250 pages of textGPT Reasoning total context window: 400KGPT Reasoning input maximum: ~680 pages of textMonthly API credits: $200 per monthly billing period (promotional OpenAI API credits for one linked API organization; they 'do not increase your ChatGPT or signed-in Codex usage limits')
Usage score
9/10 · Unlimited chat and voice subject to abuse guardrails and no five-hour limit in Work and Codex; since 29 October every Pro 200 subscription gets the higher Work and Codex allowance and the higher weekly GPT-6 Pro limit.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#5 The X Subscription Ultra xAI

Top individual tier of The X Subscription, recommended by xAI for power users. It includes everything in Super with 50x the usage of Plus, earliest access to all products and models and, per grok.com/subscription, the most powerful chat models (not named).

Price
$200.00 per month (US, excl. taxes ('Prices shown do not include applicable taxes or fees (e.g. VAT)')) $2000.00 per year (US, billed annually ($167/mo); list price excl. sales tax) 210.00 € per month (DE, regular price; page showed an intro offer '50% Rabatt für 1 Monat, danach 210 € monatlich' (105 € shown for the first month on 2026-11-01). Signed-out visitor in Germany; prices shown excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren')) 2100.00 € per year (DE, 2.100 €/Jahr (175 €/Monat), as shown on grok.com/subscription in Germany; excl. taxes ('Die angezeigten Preise verstehen sich ohne anfallende Steuern oder Gebühren'))
Models
Grok 4.7 full x.ai/pricing lists 'Latest Grok models' from Premium up; grok.com release notes (Oct 3, 2026): Grok 4.7 'now powers Expert mode on grok.com, and Fast mode for paid subscribers'. Grok Imagine (images and video) full Image and video generation included. The Imagine model per plan is not stated.
Limits
Included usage: 50x more usage compared to PlusPeak times: Priority access at peak timesEarly access: Earliest access to all products & modelsChat models: The most powerful chat models ('Die leistungsstärksten Chat-Modelle'); the models are not namedSpeed: Fastest responses and generations ('Schnellste Antworten & Generierungen')Usage pool: Every tier includes a pool of AI usage shared across Grok and Cursor. Within Grok, the pool covers Chat, Imagine, Voice, and Bot. Higher tiers include larger pools, and usage costs vary by product.Grok API: Paid plans include Grok API access. API usage is billed separately at standard model rates and doesn't draw from your plan's usage pool.Refresh: Your allowance refreshes weekly, so one heavy day never wipes out your whole month.Grok Bot: Bot usage draws from that seat’s shared weekly usage pool, not a separate additive pool.Grok Bot usage: 12x Premium usageVideo resolution: 720p videos automatically fall back to 480p once you hit the 720p cap for your tier.1080p video: Create 1080p, 15s videos (listed for Super; Ultra includes 'Everything in Super'; grok.com/subscription: 'Erstelle 1080p Videos')Beyond included usage: If you reach your included usage, you can keep going with top-ups, or wait for your allowance to refresh.Usage limits: not published (no absolute numbers)
Usage score
8/10 · 50x the Plus usage pool with priority access at peak times, refreshing weekly; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

#6 ChatGPT Pro 500 OpenAI

Highest Pro tier: the most included usage of the Pro plans (Work and Codex allowance doubled on October 29, 2026), the only one with Ultrafast mode for GPT-6 Astra and GPT-6.1 Sol in Work and Codex, and $500 a month in API credits.

Price
$500.00 per month (US, Billed monthly; annual billing and multi-month prepayment are not available)
Models
GPT-6 Astra full GPT-6 Astra in Work and Codex, including Ultrafast (Pro 500 only); pricing page: 'Pro reasoning powered by GPT-6 Astra'. The Pro tiers article refers to Pro 500's 'current weekly GPT-6 Pro message limits in Chat', while the GPT-6 help articles name only Pro $100 and Pro $200 for GPT-6 Pro in Chat. GPT-6 Sol full Powers GPT-6 Instant through Extra High in Chat; also in Work and Codex. Pricing page: 'Expanded'. GPT-6.1 Sol full ChatGPT Work and Codex only, not Chat. Pricing page: 'Expanded'. Ultrafast available on Pro 500. GPT-6.1 Terra full ChatGPT Work and Codex only, not Chat; new since Oct 26, 2026: 'available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex'. Not yet in the pricing page comparison table or the model catalog. GPT-6 Luna full Work and Codex. Pricing page: 'Expanded'. GPT-5.6 Sol full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Sol Pro full Pricing page: 'Yes'. GPT-5.6 Terra full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5.6 Luna full Work and Codex. Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' GPT-5 Thinking Mini full Pricing page: 'Unlimited*'; footnote: '*Usage must be reasonable and comply with our policies. Usage refers to Chat limits.' Legacy models full Pricing page comparison: 'Yes'.
Limits
Overall usage (Chat): Unlimited subject to abuse guardrailsIncluded usage vs. other Pro tiers: Pro 500 offers the highest included usage of the three plansUltrafast (GPT-6 Astra and GPT-6.1 Sol): Uses included usage first at 8x the Standard rate for the selected model, then the credit balance at 6x the Standard credit rateGPT-6 Pro (Chat): Weekly message limit (no number published): 'Pro 100 and Pro 500 keep their current weekly GPT-6 Pro message limits in Chat.'Voice in Chat: not published (the Voice help article lists only Pro $100 and Pro $200)Work and Codex: No five-hour usage limit in Work and Codex; the plan still has an included usage allowance (no number published)Work and Codex allowance change (Oct 29, 2026): 'Pro 100 and Pro 500 now include twice their previous Work and Codex allowance.' (no absolute number published)Deep research: Maximum deep research; usage varies by plan (no number published)GPT Instant total context window: 128KGPT Instant input maximum: ~250 pages of textGPT Reasoning total context window: 400KGPT Reasoning input maximum: ~680 pages of textMonthly API credits: $500 per monthly billing period (promotional OpenAI API credits for one linked API organization; they 'do not increase your ChatGPT or signed-in Codex usage limits')
Usage score
9/10 · OpenAI's highest included usage of the three Pro tiers and no five-hour limit in Work and Codex; no absolute numbers are published.
  • Web search
  • Deep research
  • Image generation
  • Video generation
  • Voice mode
  • File uploads
  • Coding agent
  • Browser or computer agent
  • Custom assistants or projects
  • Memory across chats

How we score

CriterionWeightWhat counts
Best included model (Text Arena)40 %The independent Text Arena score (Overall, style control) of the best model the plan includes with full access, or with limited access if none is fully included, placed between the lowest (0) and the highest (100) Arena score of all models in our catalog. Models without an Arena result give no data.
Features20 %Counted on this page: web search, deep research, image generation, video generation, voice mode, file uploads, coding agent, browser or computer agent, custom assistants or projects, memory across chats. Yes counts 1, limited 0.5, no 0, and the share of the known features gives 0–100. Unknown features are left out.
Usage allowance (editorial, 0–10)20 %Providers rarely publish exact limits, so this is an editorial score from 0 to 10 (shown as 0–100), based only on the published limits shown for each plan and compared with plans at a similar price. The reason is shown next to every score.
Monthly price in USD20 %The monthly US price when paid monthly, compared with the cheapest plan in the same price range: cheapest price ÷ this plan's price × 100, so a plan that costs 1.5 times as much gets 67. Euro prices are shown but not scored, because we never convert currencies.

Sources

  1. Plans & Pricing | Claude
  2. Get started with Claude - Claude Help Center
  3. How can I create and manage projects? - Claude Help Center
  4. Use research on Claude - Claude Help Center
  5. Can Claude produce images? - Claude Help Center
  6. Get started with Claude Motion - Claude Help Center
  7. Use voice mode - Claude Help Center
  8. Upload files to Claude - Claude Help Center
  9. Create and edit files with Claude - Claude Help Center
  10. Get started with Claude in Chrome - Claude Help Center
  11. Get started with Claude Cowork - Claude Help Center
  12. Use Claude's chat search and memory to build on previous context - Claude Help Center
  13. Claude Fable models on your plan - Claude Help Center
  14. What is the Pro plan? - Claude Help Center
  15. Use Claude Code with your Pro or Max plan - Claude Help Center
  16. Manage usage credits for paid Claude plans - Claude Help Center
  17. How large is the context window on paid Claude plans? - Claude Help Center
  18. How do usage and length limits work? - Claude Help Center
  19. Get started with Claude Duet - Claude Help Center
  20. What is the Max plan? - Claude Help Center
  21. What is the Ultra plan? - Claude Help Center
  22. Google AI Plans and Features - Gemini
  23. Google AI plans - Gemini subscriptions (Germany, German)
  24. Gemini Apps release notes
  25. Nano Banana image generation - Gemini overview (US)
  26. Gemini Apps limits & upgrades for Google AI subscribers - Gemini Apps Help
  27. Upload & analyze files in Gemini Apps - Gemini Apps Help
  28. Use Deep Research in Gemini Apps - Gemini Apps Help
  29. Gemini Omni video generation - Gemini overview (US)
  30. What is Gemini and how it works - Gemini overview (US)
  31. Ask Gemini to handle your multi-step tasks in select Android apps - Pixel Phone Help
  32. Manage your tasks & workflows in Gemini Apps - Gemini Apps Help
  33. Ask Gemini in Chrome to complete tasks for you with auto browse - Google Chrome Help
  34. Get personalization with memory of your past Gemini chats - Gemini Apps Help
  35. Limits and Plans | Jules
  36. Google AI plans with Cloud Storage - Google One (United States, English)
  37. About Gemini models - Gemini Apps Help
  38. Use Google AI Plus benefits - Google One Help
  39. Get a Google AI Plus membership - Google One Help
  40. Google One plans and pricing (Germany, German)
  41. Google AI plans with cloud storage - Google One (Germany, German)
  42. Use Google AI Pro benefits - Google One Help
  43. Manage your AI credits with Google One - Google One Help
  44. Get a Google AI Pro membership - Google One Help
  45. Use Deep Think in Gemini Apps - Gemini Apps Help
  46. Use Google AI Ultra benefits - Google One Help
  47. Get a Google AI Ultra membership - Google One Help
  48. Copilot Pricing Plans for Individuals (US)
  49. Copilot-Preispläne für Einzelpersonen (Deutschland)
  50. Conversation modes in Microsoft Copilot
  51. Introducing MAI-Image-2: for limitless creativity
  52. MAI-Image-2-Efficient: Flagship Quality, 41% Lower Cost
  53. Copilot App Updates for Personal, Work, and School Accounts
  54. What's the difference between Microsoft Copilot (free) and Copilot in Microsoft 365
  55. Deep Research in Microsoft Copilot
  56. Get Started with Create in the Microsoft Copilot App
  57. Get started with Microsoft Copilot Cowork for personal accounts (Preview)
  58. Browse with Copilot
  59. Frequently asked questions about Microsoft Copilot Notebooks
  60. Microsoft Copilot for individuals: your privacy controls and choices
  61. Compare Microsoft 365 Plans & Pricing - Microsoft Store (US)
  62. How AI usage works across Microsoft 365 subscriptions for individuals
  63. Microsoft 365-Pläne vergleichen - Microsoft Store (Deutschland)
  64. About Microsoft Copilot Pro
  65. Introducing the new Copilot with Home, Code and Autopilot - The Official Microsoft Blog
  66. Frequently asked questions about Copilot in Microsoft 365 subscriptions
  67. Introducing Microsoft 365 Premium
  68. Copilot Code (Frontier) requirements
  69. Buy Microsoft 365 Pro - Subscription Price, Download & AI Copilot | Microsoft Store
  70. Pricing | Mistral
  71. Limits and lifecycle | Mistral Docs
  72. Generate images | Mistral Docs
  73. Use Voice mode | Mistral Docs
  74. Mistral Vibe (formerly Le Chat) - AI chat and coding agent
  75. Le Chat is now Vibe | Mistral Help Center
  76. Reuse work with Skills | Mistral Docs
  77. Knowledge Base | Mistral Docs
  78. Work with Files and Canvas | Mistral Docs
  79. Search the web | Mistral Docs
  80. Group tasks with Projects | Mistral Docs
  81. Choose Work or Code | Mistral Docs
  82. Models Overview | Mistral Docs
  83. Mistral Large 4 (Le Chonk) is now open weight | Mistral
  84. Subscriptions | Mistral Docs
  85. Add context with Libraries | Mistral Docs
  86. Terms of Service for consumers in the European Economic Area
  87. Remote agents in Vibe. Powered by Mistral Medium 3.5. | Mistral AI
  88. Vibe gets to work. | Mistral AI
  89. How can I upgrade or cancel my subscription? | Mistral Help Center
  90. Spreadsheets | Mistral Docs
  91. Code Interpreter | Mistral Docs
  92. Why am I hitting API rate limits, and how do I increase them? | Mistral Help Center
  93. Pricing | ChatGPT Learn
  94. Preise | ChatGPT
  95. GPT-6 and Intelligent UI for everyone | OpenAI
  96. GPT-6 and other models in ChatGPT | OpenAI Help Center
  97. Pricing | ChatGPT
  98. ChatGPT Free Tier FAQ | OpenAI Help Center
  99. ChatGPT Voice | OpenAI Help Center
  100. What to know about the Sora discontinuation | OpenAI Help Center
  101. Using Codex with your ChatGPT plan | OpenAI Help Center
  102. ChatGPT Work and Codex | OpenAI Help Center
  103. Using cloud browser in ChatGPT | OpenAI Help Center
  104. ChatGPT agent | OpenAI Help Center
  105. Images in ChatGPT | OpenAI Help Center
  106. Deep research in ChatGPT | OpenAI Help Center
  107. Ads in ChatGPT | OpenAI Help Center
  108. What is ChatGPT Go? | OpenAI Help Center
  109. Introducing ChatGPT Go, now available worldwide | OpenAI
  110. What is ChatGPT Plus? | OpenAI Help Center
  111. Introducing GPT-6.1 Terra | OpenAI
  112. ChatGPT release notes | OpenAI Help Center
  113. About ChatGPT Pro tiers | OpenAI Help Center
  114. Managing usage with GPT-6 Astra in Work and Codex | OpenAI Help Center
  115. ChatGPT Plan | Pro
  116. API credits benefit for eligible ChatGPT subscriptions | OpenAI Help Center
  117. Getting started with your dot | OpenAI Help Center
  118. Perplexity Pricing: Plans for Individuals and Enterprise
  119. What advanced AI models are included in my subscription? - Perplexity Help Center
  120. Which Perplexity Subscription Plan is right for you? - Perplexity Help Center
  121. File Uploads - Perplexity Help Center
  122. Generating Images with Perplexity - Perplexity Help Center
  123. How credits work on Perplexity - Perplexity Help Center
  124. Perplexity Hub (FAQ: How much does it cost?)
  125. What is Research mode? - Perplexity Help Center
  126. Memory - Perplexity Help Center
  127. How to use the Perplexity Voice Assistant for iOS - Perplexity Help Center
  128. What is Model Council? - Perplexity Help Center
  129. What is Pro Search? - Perplexity Help Center
  130. What is Computer? - Perplexity Help Center
  131. Using the Connector for Slack - Perplexity Help Center
  132. What is Perplexity Pro? - Perplexity Help Center
  133. Generating Videos with Perplexity - Perplexity Help Center
  134. Perplexity Pro
  135. What's New in Advanced Deep Research - Perplexity Help Center
  136. What are Projects? - Perplexity Help Center
  137. Paying in Local Currency - Perplexity Help Center
  138. Effort Mode, GPT-6 Astra, and Skills Marketplace - Perplexity Changelog
  139. Ultrafast Computer models, Decisions API, and connector controls - Perplexity Changelog
  140. Automations, inline visualizations, and GPT-6.1 Sol - Perplexity Changelog
  141. What is Brain? - Perplexity Help Center
  142. Perplexity Max - Perplexity Help Center
  143. Pricing: Compare Plans | SpaceXSI
  144. Grok Build | SpaceXSI
  145. Introducing The X Subscription (Oct 12, 2026)
  146. FAQ - Grok Website / Apps | SpaceXSI Docs
  147. Grok — Useful AI Chatbot with Voice & Image Generation | SpaceXSI
  148. What is The X Subscription? | SpaceXSI Docs
  149. Pricing (The X Subscription) | SpaceXSI Docs
  150. Grok plan picker (The X Subscription, EUR prices shown in Germany)
  151. Usage limits | SpaceXSI Docs
  152. Grok Bot on The X Subscription | SpaceXSI Docs
  153. Grok Release Notes (grok.com)
  154. X Premium Customers moving to The X Subscription | SpaceXSI Docs
  155. Grok Bot | SpaceXSI Docs
  156. AI teammates that finish the work | Grok Bot