Skip to main content
Back to OneHuman

AI Tool Pricing Compared

💡 All pricing and data verified by OneHuman or community. Last updated: October 31, 2026

ChatGPT

OneHuman Verified
$8–$500Free Version:
Monthly Price:$8–$500
Free Version:
Education Pricing:ChatGPT Edu (universities only, via sales). No student discount
Latest Model:GPT-6 Sol in Chat (Plus/Pro/Business/Enterprise, Oct 7 2026; Instant to Extra High); GPT-6 Luna in Chat (Free/Go, from Oct 8); Pro reasoning option = GPT-6 Astra. GPT-6.1 Sol (Sep 29) and GPT-6.1 Terra (Oct 26, Plus and above) Work/Codex only. GPT-5.5 retired from Codex (ChatGPT sign-in) Oct 14.
Context Window:54K/256K (Instant/Reasoning, Go+Plus) / 128K/400K (Instant/Reasoning, Pro, all 3 tiers)
Daily/Message Limits:Free: unlimited text chats with GPT-6 (GPT-6 Luna; subject to abuse guardrails, OpenAI pricing page, Oct 30); Go: more messages with tools than Free (GPT-6 Luna, may include ads; no message count published); Plus: OpenAI publishes no fixed message count for Plus; limits depend on the model and thinking level, and ChatGPT shows the reset time (OpenAI help page, checked Oct 20); Pro $100 (5x); Pro $200 (allowances changing for new subscribers, OpenAI Sept 29); Pro $500 (new tier, Sept 29 2026, highest usage) — Astra via the Pro reasoning option on all 3 Pro tiers; Pro includes 1 Dot (always-on agent, Sept 29 2026, eligible markets, extra Dots priced later); GPT-6.1 Sol and GPT-6.1 Terra (Oct 26; API $1/$5 per 1M tokens) are Work/Codex only, not in Chat on any tier
User Base:800M+
Specialization:General-purpose AI assistant
Top Feature:GPT-6 with Intelligent UI in Chat (text + visuals + interactive elements) on every plan since Oct 7-8; Astra via the Pro reasoning option
Top Limitation:Peak hour usage caps
Best For:General tasks & plugins

Muse

OneHuman Verified
$0–$100Free Version:
Monthly Price:$0–$100
Free Version:
Education Pricing:None found
Latest Model:Muse Spark (agent, consumer app); API: Muse Spark 1.3 (1M context)
Context Window:Undisclosed (app); 1M (API)
Daily/Message Limits:Free: usage limit, ~100M tokens/wk (per Zuckerberg); Power $20/mo: 500M tokens/wk; Maximum $100/mo: 3B tokens/wk. In limited testing; not available in all locations yet (Meta).
User Base:1.8M+ downloads (Apptopia est.)
Specialization:Autonomous agent: acts on your behalf
Top Feature:New Sept 8 2026; own persistent cloud VM + browser; Power $20 = 500M tokens/wk
Top Limitation:Limited testing, not in all locations; tokens/week allowance; context undisclosed
Best For:Delegating errands & bookings

Claude

OneHuman Verified
$20–$200Free Version:
Monthly Price:$20–$200
Free Version:
Education Pricing:Claude for Education (universities only, via sales). No student discount
Latest Model:Sonnet 5.5 (Sep 28 2026), Opus 5.5 (Sep 22 2026, $4/$20/MTok, down from $5/$25), Fable 5.1 (Sep 1 2026, cache reads cut 75% to $0.25), Haiku 5.5 (released Oct 7 2026, $0.10/$0.50/MTok for prompts up to 100K, 5x above). Sonnet 5.5 cache reads halved to $0.10/MTok (Oct 7)
Context Window:Up to 1M (varies by model)
Daily/Message Limits:≥5x Free per 5h session (Anthropic states no fixed message count); Max $100 (5x Pro per 5h and per week); Ultra $200 (10x Pro per week, no 5-hour limit for verified accounts; Anthropic pricing page, Oct 31 — the Oct 30 page listed this tier as Max 20x); Fable included up to half the weekly limit on Max/Ultra, via usage credits on Pro; Opus 5.5 launch increased 5h limits on Pro/Max/Team/Enterprise (no published figure)
User Base:19M+
Specialization:Long-form content & analysis
Top Feature:Sonnet 5.5 (Sep 28), Opus 5.5 (Sep 22, $4/$20/MTok, down from $5/$25), Fable 5.1 (Sep 1, cache reads -75%) — 1M context; Haiku 5.5 released Oct 7 ($0.10/$0.50, 5x above 100K prompts); Sonnet 5.5 cache reads halved to $0.10
Top Limitation:Slower response times
Best For:Writing & deep analysis

Gemini

OneHuman Verified
$4.99–$199.99Free Version:
Monthly Price:$4.99–$199.99
Free Version:
Education Pricing:Free 1 year for students (US: AI Pro; elsewhere: AI Plus). Card required; auto-renews after
Latest Model:Gemini 4 Argon (flagship; AI Pro and Ultra, generally available Oct 13 2026); Auto (free tier, mostly Flash-Lite per Google, since Oct 9 2026); Flash on AI Plus
Context Window:model-dependent (3.x gen)
Daily/Message Limits:Free: Auto model selection, mostly Flash-Lite (Google), 32k context; Google AI Plus $4.99/mo (Flash-Lite and Flash, no Pro since Oct 2026; 128k context); Pro $19.99/mo (Gemini 4 Argon, Deep Research); Ultra $99.99/mo (5x) or $199.99/mo (20x), highest Argon access. Argon API: $2/$10 per 1M tokens through Nov 30 2026, $4/$20 from Dec 1. Refreshes every 5h up to a weekly cap; AI credits purchasable.
User Base:450M+
Specialization:Multimodal & Google integration
Top Feature:Gemini 4 Argon (AI Pro and Ultra, since Oct 13 2026); API intro price $2/$10 ends Nov 30
Top Limitation:Frequent capability changes
Best For:Google Workspace users

Mistral

OneHuman Verified
$14.99Free Version:
Monthly Price:$14.99
Free Version:
Education Pricing:$5.99/mo for verified students (normally $14.99)
Latest Model:Medium 3.5 (default), Small 4 (new), Magistral Medium/Small (reasoning); Large 4 (public preview Oct 6 2026, API only via Mistral Studio; open weights due end of October)
Context Window:256K (Med 3.5)
Daily/Message Limits:Free: limited messages and web searches (no figure published). Pro: up to 6x Free's messages, up to 5x its web searches (Mistral pricing page, Oct 30)
User Base:3M+
Specialization:European privacy & code
Top Feature:Medium 3.5 default; Large 4 preview (1T params, 49B active, 160+ languages, trained in Mistral's European datacenters); + Small 4 ($0.10/$0.30), Magistral Medium/Small (reasoning)
Top Limitation:Smaller user base
Best For:Privacy-conscious teams

Perplexity

OneHuman Verified
$17–$167Free Version:
Monthly Price:$17–$167
Free Version:
Education Pricing:Education Pro $9/mo billed annually (verified students and educators)
Latest Model:Model picker incl. Claude Sonnet 5, GPT-5.6, Gemini 3.1 Pro, Sonar 2
Context Window:Model-dependent
Daily/Message Limits:Pro $17/mo annual (Perplexity Computer, advanced models); Max $167/mo annual (unlimited, priority). Credits: Pro 4,000 bonus credits (+$40 free Computer credits, limited time); Max 10,000 monthly + 35,000 bonus credits (+$450, limited time)
User Base:20M+
Specialization:Real-time research
Top Feature:Gemini 3.1 Pro, Sonar 2, Claude Sonnet 5 — model picker + unlimited Pro searches
Top Limitation:Search-focused only
Best For:Research & fact-checking

Grok

OneHuman Verified
$8–$200Free Version:
Monthly Price:$8–$200
Free Version:
Education Pricing:None
Latest Model:Grok 4.7 (Sep 21 2026, flagship — supersedes 4.6)
Context Window:500K (Grok 4.7)
Daily/Message Limits:The X Subscription (Oct 12 2026): Free limited; Plus $8 base pool; Premium $30 = 4x Plus; Super $100 = 15x; Ultra $200 = 50x. One weekly pool shared by Grok and Cursor; base pool size not published. Promo seen Oct 30: 50% off the first month on Premium/Super/Ultra (list prices unchanged). Old SuperGrok/Heavy plans kept only for existing subscribers. SpaceX-owned (brand: SpaceXSI).
User Base:30M+
Specialization:Uncensored real-time reasoning
Top Feature:Grok 4.7 flagship (Sep 21 2026, API $2/$6) + one plan includes Grok, Cursor and X Premium
Top Limitation:Shared Grok/Cursor usage pool with undisclosed size; non-Grok models in Cursor cost extra
Best For:News & social media trends

GitHub Copilot

OneHuman Verified
$10–$100Free Version:
Monthly Price:$10–$100
Free Version:
Education Pricing:Copilot Student: free for verified students and teachers (Auto model selection only)
Latest Model:Model choice varies by plan: Haiku 4.5 and GPT-5 mini on Free, premium models incl. Opus on Pro+ (GitHub plans page, Oct 30)
Context Window:Not stated by GitHub
Daily/Message Limits:AI Credits: 1 credit=$0.01. Free: 2,000 completions/mo; Pro $10 ($15 monthly credits); Pro+ $39 ($70); Max $100 ($200). New signups unfrozen.
User Base:100M+
Specialization:AI pair programming & code completion
Top Feature:IDE code completion & chat (VS Code, JetBrains); Max tier $100/mo added Jun 2026; AI Credits model
Top Limitation:Multi-file context blindness in large codebases
Best For:Developers in VS Code/JetBrains IDEs

DeepSeek

OneHuman Verified
FreeFree Version:
Monthly Price:Free
Free Version:
Education Pricing:Free for everyone (no paid tier)
Latest Model:DeepSeek V4-Pro-0813 / V4.1-Flash-1017 (deepseek-flash)
Context Window:1M ctx, 384K max output (both)
Daily/Message Limits:No cap (server load varies). API Search/Fetch tool calls, including calls made through Claude Code, Codex or DeepSeek Harness, are billed $0.75 per 1,000 requests each from Nov 1 2026 (free until then)
User Base:131M+
Specialization:Code & math reasoning
Top Feature:V4-Pro-0813 / V4.1-Flash, 1M context, 384K max output. V4.1-Flash $0.15/$0.60 off-peak or $0.30/$1.20 peak per 1M tokens (in/out, retired V4-Flash Sept 2026)
Top Limitation:Primarily code-focused
Best For:Developers & engineers

💡 All pricing and data verified by OneHuman or community. Last updated: October 31, 2026

Quick answers

Quick answers, from the table below

ChatGPT Plus vs Claude Pro vs Gemini AI Pro: price and limits

Price: ChatGPT Plus $20/month, Claude Pro $20/month, Google AI Pro $19.99/month.

Limits: OpenAI publishes no fixed message count for ChatGPT Plus; limits depend on the model and thinking level, and ChatGPT shows the reset time. Claude Pro is at least 5x the Free allowance per 5-hour session, and Anthropic states no fixed message count. Google AI Pro refreshes every 5 hours up to a weekly cap, with AI credits purchasable for more.

Context window: ChatGPT Plus 54K (Instant) / 256K (Reasoning). Claude up to 1M (varies by model). Gemini is model-dependent; Google AI Pro includes Gemini 4 Argon since Oct 13 2026.

What do Claude Max and Claude Ultra add over Claude Pro at $100 and $200 per month?

Claude Max is $100/month (5x Pro usage per 5 hours and per week); Claude Ultra is $200/month (10x Pro usage per week, no 5-hour limit for verified accounts). Claude Pro is $20/month. Anthropic's pricing page listed the $200 plan as Max 20x on Oct 30 and as Ultra on Oct 31 2026.

Since Oct 7 2026, Max and Team also include a monthly API credit. It does not cover interactive Claude Code.

Anthropic states no fixed message count for any plan, so the multipliers are relative allowances, not message totals.

ChatGPT Pro vs Claude Max: tiers, prices and published limits

ChatGPT Pro has three tiers since Sep 29 2026: $100 (5x), $200 and $500 per month. For the $200 tier OpenAI says included allowances are changing for new subscribers, and no first-party multiplier is published. Pro context is 128K / 400K (Instant / Reasoning).

Claude has a $100 Max tier (5x Pro) and a $200 Ultra tier (10x Pro weekly, no 5-hour limit for verified accounts). Claude context is up to 1M (varies by model).

Gemini sits alongside: Google AI Ultra is $99.99/month (5x) or $199.99/month (20x). It went from a single $249.99/month tier (plans page archived May 17, 2026) to $99.99 and $199.99 tiers at I/O on May 19, 2026.

Does the price of these plans change by country?

For ChatGPT, the tracked prices are $20.00/mo in the US, ₹399/mo (~$4.60) — 77% off in India, $4.99/mo — 75% off in Argentina and ~$13.60/mo — 18% off in Turkey. Other tools list their own regional prices in the Pro Deep Analysis table below.

GitHub Copilot
OneHuman Verified

$10/month
Free Version Available

Key Features

  • AI code completion and chat across IDEs (VS Code, JetBrains, Visual Studio, Neovim)
  • Multi-model chat picker (GPT-5 family, Claude, Gemini), CLI (gh copilot), and PR description generation
  • Plans: Pro $10/mo ($15 monthly credits), Pro+ $39/mo ($70), Max $100/mo ($200; new Jun 2026) — AI Credits model (1 credit = $0.01)

User Considerations

  • WARNING: Claude Opus access removed from Pro ($10) plan Apr 20 2026; Pro+ ($40) Opus 4.7 quota cut 60% via 7.5x multiplier; refunds available until May 20 2026
  • Multi-file context blindness in coding workflows — Copilot frequently loses track of cross-file dependencies in larger codebases; agentic tasks complete noticeably slower than rivals
✅
GitHub Copilot pricing verified
Last verified: October 31, 2026
GitHub Copilot limits and pricing, plan by plan →

Mistral
OneHuman Verified

$14.99/month
Free Version Available

Key Features

  • Mistral Medium 3.5 (128B, default, 77.6% SWE-bench); + Small 4 (new, $0.10/$0.30/MTok); Magistral Medium & Small (new reasoning models, Jun 2026)
  • Canvas, code interpreter, web search, image creation; Vibe remote agent platform launched Apr 29 2026
  • 256K context window (Medium 3.5), API: $1.5/M input tokens / $7.5/M output tokens
  • European-based models, technical appeal, modified MIT license, No Telemetry Mode for confidential data

User Considerations

  • Le Chat interface and ecosystem less polished and feature-rich than larger assistants
  • Non-expert buyers may find API model and pricing choices more complex than single-plan offerings
  • Knowledge freshness can lag, with occasional out-of-date information
  • Integrations into mainstream productivity tools still developing, requiring custom engineering
  • Support and documentation aimed at technical teams, barrier for non-technical organizations
  • Some users report nonsensical or incoherent outputs on complex reasoning tasks — a quality consistency flag beyond the knowledge-freshness issue
  • Until the weights ship, a vetted group (cybersecurity leaders, partners, state authorities) gets Large 4 with reduced moderation and expanded cyber capabilities; the public preview is the moderated version.
✅
Mistral pricing verified
Last verified: October 31, 2026
Mistral limits and pricing, plan by plan →

Gemini
OneHuman Verified

$4.99/month
Free Version Available

Key Features

  • Gemini 4 Argon is the flagship in the Gemini app for AI Pro and Ultra (generally available since Oct 13 2026); API intro price $2/$10 per 1M tokens through Nov 30 2026, then $4/$20 from Dec 1
  • Free tier runs on Auto model selection, mostly Flash-Lite per Google (since Oct 9 2026; 32k context); Gemini 3.8 Flash (paid/API, Sept 2 2026) and 3.7 Flash at the same price — API gen 3.x launched June 2026
  • Deep Workspace integration (Docs, Sheets, Gmail, Drive)
  • Skills (Gems auto-migrate Nov 2026), Canvas, audio summaries
  • Multimodal (text+image), "Google it" verification button
  • Google AI tiers: Plus $4.99/mo (Flash-Lite and Flash, no Pro since Oct 2026; 128k context), Pro $19.99/mo (Gemini 4 Argon, Deep Research), Ultra $99.99/mo (5×) or $199.99/mo (20×) — usage refreshes every 5h up to a weekly cap, AI credits purchasable for more

User Considerations

  • Gemini 4 Carbon (announced Oct 28 2026) is limited to a subset of cyber defenders and is not available to the public yet
  • Some users report session instability in Pro/3.x, including chat resets or lost history
  • More prone to hallucinations despite verification
  • Usage limits and model options (Free, Pro, Ultra, Workspace) can be confusing
  • Safety policies around sensitive topics can feel stricter than other tools
  • Support and billing resolution may feel slow when Pro features change unexpectedly
  • Responses tend toward verbose, bullet-heavy formatting even when prose would be clearer — one of the most consistent style complaints in 2026 reviews
✅
Gemini pricing verified
Last verified: October 31, 2026
Gemini limits and pricing, plan by plan →

Grok
OneHuman Verified

$30/month
Free Version Available

Key Features

  • X/Twitter data access for real-time discourse and memes
  • Fast conversational assistant, distinct personality
  • Grok 4.7 (Sep 21 2026, flagship, 500K context, API $2/$6 under 200K); Grok 4.3 remains the cheapest tier. The X Subscription (Oct 12 2026) bundles Grok, Cursor and X Premium: Free, Plus $8, Premium $30 (4x Plus), Super $100 (15x), Ultra $200 (50x), Teams from $30/seat.
  • Awareness of current trends and memes

User Considerations

  • Depth and accuracy on complex tasks can be less consistent than mature assistants
  • Strongest experience is tied to the X ecosystem, limiting for non-X users
  • Feature availability varies by region and device, rougher for non-iOS/non-US users
  • Integration options beyond X are still relatively limited
  • Conversational tone may feel less neutral, a drawback for formal or professional use
  • Documented 2026 safety incidents involving inappropriate content generation — relevant for professional or brand-sensitive use cases where output reliability matters
  • xAI merged into SpaceX (Feb 2026) — Grok user data now sits inside a defense contractor's corporate structure. Data-flow boundary between the Grok consumer product and SpaceX operations has not been published. EU GDPR data controller post-merger is unclear. Regulatory investigations open in EU, UK, and US.
  • Community reports of intermittent access issues — lost conversation history and desktop-app instability described in ongoing user threads, not yet acknowledged by xAI
  • Pay-as-you-go credit top-ups for image/video generation criticized as poor, opaque value — users report $20 in credits yielding roughly half that in usable output, with no published credit-to-generation conversion rate
  • Shared pool: heavy Cursor use draws down your Grok chat, image and voice allowance and vice versa; xAI publishes only multiples of the Plus pool, not the base allowance
  • Claude, GPT and other non-Grok models inside Cursor are billed at API list price on top of the plan, Premium and above only
✅
Grok pricing verified
Last verified: October 31, 2026
Grok limits and pricing, plan by plan →

ChatGPT
OneHuman Verified

$8/month
Free Version Available

Key Features

  • Go tier ($8/mo): GPT-6 Luna with ads, more messages than Free — the cheapest entry to paid ChatGPT
  • GPT-6 Sol in Chat on Plus/Pro/Business/Enterprise (Oct 7 2026), GPT-6 Luna on Free/Go, GPT-6 Astra via the Pro reasoning option. GPT-6.1 Sol (Sep 29 2026) is Work/Codex only, never in Chat, DALL-E 3, GPT Store with custom GPTs
  • Canvas for collaborative editing, Projects, voice mode
  • 54K/256K context (Instant/Reasoning, Go/Plus) / 128K/400K (Instant/Reasoning, Pro) — Pro now has 3 tiers: $100, $200, and $500 (new Sep 29 2026)
  • Pro includes 1 Dot (always-on agent; Sep 29 2026; eligible markets; extra Dots to be priced later)
  • Integrated web search and mature ecosystem

User Considerations

  • Quality regression reports with recent models
  • Over-cautious safety filters and moralizing tone
  • Usage caps and throttling can appear unexpectedly, even on paid plans
  • Context amnesia in long projects
  • Billing and feature tiers (Go, Plus, Pro) can feel complex to compare
  • Cancellation window closes 24 hours before renewal; refunds reported as difficult outside EU — check your jurisdiction before subscribing
  • A federal court order in ongoing NYT v. OpenAI litigation has required OpenAI to preserve all ChatGPT conversation logs indefinitely, including chats users deleted — a real limit on the platform's stated deletion policy users should know about
✅
ChatGPT pricing verified
Last verified: October 31, 2026
ChatGPT limits and pricing, plan by plan →

DeepSeek
OneHuman Verified

$0/month
Free Version Available

Key Features

  • API Search and Fetch tool calls (also when made through Claude Code, Codex or DeepSeek Harness) are free until Nov 1 2026, then billed at $0.75 per 1,000 successful requests each
  • DeepSeek V4-Pro-0813 and V4.1-Flash (V4-Flash retired, Sept 2026) — 1M token context window, 384K max output, both
  • V4.1-Flash: $0.15/$0.60 off-peak or $0.30/$1.20 peak per 1M tokens; V4-Pro: $0.66/$1.98 off-peak or $1.32/$3.96 peak — still among the cheapest APIs
  • Strong code and math/logic performance, open-source friendly, no paid subscription tier
  • Research-grade models, cost-quality ratio

User Considerations

  • Main experience is developer- and API-centric; non-technical users may find it less accessible
  • Interface and documentation described as spartan compared to US/EU incumbents
  • Free endpoints experience load-related slowdowns and soft caps limiting heavy use
  • Integrations into Western productivity stacks still limited, often requiring custom work
  • Data governance and regulatory questions due to vendor's jurisdiction require extra due diligence
  • Post-update quality regression reported by users — some describe V3→V4 updates as producing less nuanced responses on complex tasks
  • Security vulnerability reports on sensitive prompts surfaced in 2026; treat as unsuitable for confidential inputs regardless of hosting status
✅
DeepSeek pricing verified
Last verified: October 31, 2026
DeepSeek limits and pricing, plan by plan →

Perplexity
OneHuman Verified

$17/month
Free Version Available

Key Features

  • Real-time web search with strong citations
  • Model picker: Claude Sonnet 5, GPT-5.6 Sol, Gemini 3.1 Pro, Sonar 2 + Mistral, DeepSeek
  • Focus modes (writing, academic, video, social, math)
  • Unlimited Pro searches, Deep Research, Collections, Threads, Spaces
  • Pro $17/mo (annual billing), Max $167/mo annual: unlimited search, advanced reasoning models, priority access

User Considerations

  • Answers feel surface-level despite many sources
  • Free tier's Deep/Pro search limits reached quickly by heavy researchers
  • Changing caps and experiments with ads create uncertainty about long-term value
  • Interface and result layout can feel cluttered vs simpler chat assistants
  • Features and behaviours can differ across web, mobile, and extensions
  • Local and geographic search quality lags behind general web search; coverage gaps for non-English queries and regional topics
✅
Perplexity pricing verified
Last verified: October 31, 2026
Perplexity limits and pricing, plan by plan →

Claude
OneHuman Verified

$20/month
Free Version Available

Key Features

  • Claude Sonnet 5.5 (Sep 28 2026, $2/$10/MTok, same as Sonnet 5), Opus 5.5 (Sep 22 2026, $4/$20/MTok, down from $5/$25), Fable 5.1 (Sep 1 2026, $10/$50/MTok, cache reads cut 75% to $0.25) — up to 1M context (varies by model). Mythos 5.1 is the same model as Fable 5.1 with different safeguards, restricted to Anthropic's trusted-access programs.
  • 1M token context window (GA since Mar 13 2026 on all plans), Artifacts, Projects
  • Extended thinking mode, code assistant
  • Strong long-document and research handling
  • Max $100/mo (5x Pro usage per 5h and per week) and Ultra $200/mo (10x Pro usage per week, no 5-hour limit for verified accounts) for heavy use — Pro $20/mo remains the standard entry tier
  • Monthly API credit (Oct 7 2026): Max $100, Ultra $200; Team pooled ($20 per Standard seat, $100 per Premium seat), capped at $500. Free, Pro and Enterprise: none.

User Considerations

  • Surprise unannounced usage limit reductions reported on higher-priced tiers
  • Complex interactions consume limits quickly, especially with large files or extended thinking
  • Communication around limit changes and roadmap decisions could be clearer
  • Fewer third-party integrations and plugins than OpenAI or Microsoft
  • Cautious style can feel overly verbose or hedged, slowing decision-making
  • 3 partial outages recorded in April 2026 — including one on the same day as a major model launch; no published uptime SLAs for consumer or API tiers
  • The credit is for the Claude Platform (API, Batches, Console, Managed Agents, Agent SDK with a Console key). It does not cover interactive Claude Code (terminal, IDE, desktop, web), extra usage, or Claude on Bedrock/Vertex/Foundry. No rollover; one linked Console organization you cannot change yourself.
  • Anthropic's May 13 help page promised the credit to Pro, Max, Team and Enterprise; Pro and Enterprise did not get it (see OneHuman's BNA32 report, which cites the June 4 Wayback capture).
  • Haiku 5.5's "around 75% less" is an average: every price line is 5x higher once a prompt passes 100K tokens.
  • Opus 5.5 launch limit reset (Sep 22 2026): given occasionally to undefined eligible plans; an unused reset can expire and is lost on downgrade or cancel; it cannot be redeemed from Claude Code or mobile.
✅
Claude pricing verified
Last verified: October 31, 2026
Claude limits and pricing, plan by plan →

Muse
OneHuman Verified

$20/month
Free Version Available

Key Features

  • Personal AI agent, not just chat: takes actions (email, calendar, shopping, bookings) with your approval, with a full audit trail
  • Your own persistent cloud computer: each user gets a private Linux VM with a full browser the agent works in, and you can step in at any time
  • Works where you already are: iOS, Android, Mac, web (muse.ai) and inside WhatsApp
  • Powered by Muse Spark. 3 tiers: Free (usage limit, ~100M tokens/wk per Zuckerberg), Power $20/mo (500M tokens/wk), Maximum $100/mo (3B tokens/wk)
  • Checkout via Link (Stripe) with a one-time card number, so the merchant and the agent don't see your real card (Meta's claim)

User Considerations

  • An agent that can spend money and send messages on your behalf raises the stakes of mistakes. Meta's own VP said "it is impossible to say that there is never going to be a mistake" (GuruFocus/Yahoo, Sept 9). Keep approvals on for purchases
  • Meta states Muse conversations and VM data aren't shared with its ad systems. This is Meta's claim, not independently verifiable
  • Launched 13 days after Meta's $18B multistate child-safety settlement (CNN, Aug 26–27, 2026). Relevant for families weighing an agent with account access
  • The cheapest API tier ("Contributor", $0.10/$0.20) lets Meta use your prompts and outputs to improve its products. The discount is paid for with data
  • A payment card is reportedly required even on the free tier, per multiple independent sources — not confirmed on Meta's own pages
  • Brand-new product (Sept 8, 2026). Tiers, limits and availability are likely to shift, so recheck every cycle
✅
Muse pricing verified
Last verified: October 31, 2026
Muse limits and pricing, plan by plan →

Deeper analysis

What the marketing pages don't tell you — rate limits, context windows, API costs, regional pricing

Pro
Rate LimitsContext WindowsAPI CostsRegional Pricing
ChatGPTUnlimited text chats with GPT-6 Luna since Oct 8 (subject to abuse guardrails; OpenAI pricing page, Oct 30)OpenAI publishes no fixed message count for Plus; limits depend on the model and thinking level, and ChatGPT shows the reset time (OpenAI help page, checked Oct 20); GPT-6 Sol in Chat since Oct 7 (Plus/Pro/Business/Enterprise)Go ($8) gets more messages with tools than Free and may include ads (GPT-6 Luna); OpenAI publishes no message count.
ClaudeLimited usage every 5 hours and every week (Anthropic states no fixed message count)≥5x Free per 5hr session (Anthropic states no fixed message count)Sonnet 5.5 (Sep 28 2026), Opus 5.5 (Sep 22, $4/$20/MTok, down from $5/$25), and Fable 5.1 (Sep 1, cache reads cut 75%) are the current…
GeminiAuto model selection, mostly Flash-Lite per Google (since Oct 9 2026); 32k context; compute-based limitsRefreshes every 5h up to a weekly cap; AI credits purchasable for moreFree tier runs on Auto since Oct 9 2026 (mostly Flash-Lite per Google; Plus gets Flash-Lite and Flash, no Pro), not 3.8 Flash — 3.8…
GitHub Copilot2,000 completions/mo + 50 chat requests (Haiku 4.5, GPT-5 mini)Pro $10 ($15 monthly credits), Pro+ $39 ($70), Max $100 ($200) — AI Credits model (1 credit = $0.01)Free tier limits and included models confirmed direct from github.com/features/copilot/plans this cycle (Aug 2026) — previously undisclosed.
See the full comparison — every column, every tool, updated every 15 days. Pro $4.95/month →