Perplexity Personal Computer vs OpenAI Codex Computer Use vs Claude Computer Use: Which AI Should Run Your Mac?

April 2026 is the month computer-use stopped being a tech demo. Three vendors -- Perplexity, OpenAI, and Anthropic -- now each ship a real product where an AI can move your cursor, type into your apps, click around your browser, and finish work for you. They look superficially similar. They are not interchangeable.

Each one is built around a different theory of where the agent should live and who should be in the loop. Perplexity bets on the always-on Mac. OpenAI bets on the work-thread inside Codex. Anthropic bets on the developer who wants to wire computer-use into their own workflow. This guide lays out who each one is for, where each one wins, and how to pick.

Key Takeaways

• Perplexity Personal Computer is best if you want a 24/7 Mac mini agent and already pay for Perplexity Max ($200/mo).
• OpenAI Codex Computer Use is best if your work is already inside Codex and you want the agent to drive code, browser, and Mac apps without leaving the thread.
• Claude Computer Use is best if you want to wire computer-use into your own scripts via API, or switch on the built-in version in Claude Code and Claude Desktop (macOS and Windows, Pro or Max plan), which approves apps one at a time.
• All three can read your screen and take actions; none watch you when not running a task.
• Pricing is wildly different: Perplexity is gated to a $200/mo tier; Codex piggybacks on a $20-200 ChatGPT plan; Claude is metered API tokens, or a Pro or Max subscription inside Claude Code.
• OS support has changed since launch. All three arrived Mac-first in April 2026, but Codex Computer Use added Windows desktop apps in late May 2026 and OpenAI now documents it on macOS and Windows. Claude Computer Use is OS-agnostic via API, is the only one that runs on Linux, and Claude Desktop ships it on macOS and Windows.

If you want these comparisons each week, subscribe here.

The 30-Second Decision

If you only have 30 seconds, answer these three questions in order and stop at the first "yes":

1. Is your work mostly coding inside ChatGPT or Codex? Choose OpenAI Codex Computer Use. It shares context with the agent already writing your code. On Windows, Codex and Claude Desktop both have first-party hosts; Perplexity does not.
2. Do you want a turnkey Mac app with zero setup, and can you justify $200/month? Choose Perplexity Personal Computer. It is the most polished cross-app personal agent, gated behind Perplexity Max.
3. Are you building computer-use into a product, a CI pipeline, or a Linux environment? Choose Claude Computer Use. It is the only one of the three that runs anywhere through an API, on any OS.

If you answered no to all three, default to Codex Computer Use -- it rides on a subscription most readers already hold. The rest of this guide is the long version of that decision.

Codex Computer Use vs Claude Computer Use: The Straight Answer

These two get compared against each other more than either gets compared to Perplexity, and the honest answer is that they are not really competing for the same buyer.

Choose Codex Computer Use if you want to use a computer-use agent. It is a finished product: install the ChatGPT desktop app, add the Computer Use plugin, grant Screen Recording and Accessibility, and the agent drives your apps. You get the in-app browser, DOM-precise context capture, and shared state with the agent writing your code. You do not build anything.

Choose Claude Computer Use if you want to build with a computer-use agent, or if you already work in Claude Code. Through the API it is a model toolset: you send a screenshot, it returns an action, you execute the action, you loop. You decide the host, the sandbox, and the audit trail -- which is why it is the only one of the three that can run inside a Docker container in CI. Inside Claude Code it is a built-in server you switch on (the CLI on macOS, Claude Desktop on macOS and Windows), a research preview for Pro and Max plans that approves apps one at a time and stops on Esc.

The differences that actually decide it:

Two practical notes that catch people out. First, neither one is unattended infrastructure by default: Codex needs a desktop session on a machine that is awake and signed in, and Claude needs whatever uptime your own harness has. Second, "Windows support" is not one thing -- Codex runs on Windows, but only on the active desktop, so a Windows machine cannot be both your workstation and your agent host at the same time. If the leave-it-running pattern is the reason you want this, that pushes you to a Mac for Codex, or to Claude with your own host.

If you are choosing between all three rather than just these two, the full side-by-side table further down is the wider view.

What Counts as "Computer Use" in 2026

A computer-use agent is one that takes actions through the same interface a human would: it sees the screen, moves the mouse, clicks buttons, types into fields, and reads what came back. The category started in late 2024 with Anthropic's first computer-use beta. By April 2026 it has split into three meaningfully different shapes:

• A screen agent sits on your machine and operates whatever is on your active display. It works across apps because it works at the same layer the human does.
• A browser-scoped agent works inside a built-in browser pane and captures the DOM directly, which makes it more precise but limited to web pages.
• An API agent is a model endpoint that takes a screenshot in and returns an action out. The developer wires it into whatever loop they want.

Perplexity Personal Computer is the first kind. OpenAI's Codex Computer Use is a hybrid of the first and second kinds. Claude Computer Use is fundamentally the third kind, with packaging on top from Anthropic and from third parties.

Perplexity Personal Computer: The Always-On Mac Agent

Perplexity launched Personal Computer on April 16, 2026. It runs as a Mac app that hooks into local files, native macOS apps (Mail, iMessage, Calendar, Notes, Files), and your browser, and presents a single unified prompt surface activated by pressing both Command keys. Perplexity recommends running it on a dedicated Mac mini so the agent can stay on 24/7 -- the marketing line is "always on, always working". For setup details, pricing, and the Mac mini workflow pattern, use the Perplexity Personal Computer guide.

The product has three structural decisions worth understanding before you buy it.

First, Perplexity owns the orchestration layer. The agent does not just run a single tool call -- it composes work across your apps. Ask it "find the contract Sara sent last week and summarize the payment terms" and it can search Mail, open the PDF in Files, summarize, and reply with the answer in a single turn. That is qualitatively different from an agent that only runs in a browser tab.

Second, everything runs in a sandbox with a kill switch. Per Perplexity's own framing, every action is auditable and reversible, and there is a visible button to halt the agent mid-action. This is the right answer for the threat model -- once an agent can drive your computer, you need a way to slam the brakes.

Third, the price tag is high. Personal Computer is a Perplexity Max-only feature at $200/month, with a waitlist for everyone else. If you do not already have Max, decide whether the Personal Computer setup and Mac mini workflow justify the upgrade from the $20 Pro plan.

How a reader actually uses Perplexity Personal Computer

A few concrete patterns from week one of the launch:

• Inbox triage. "Read the last 50 emails, list anything that needs a reply today, and draft a one-line response for each." Personal Computer reads Mail, ranks by urgency, and writes drafts you approve.
• Meeting prep. "For my 2pm call with Acme, pull our last three email threads, the most recent contract draft, and any Slack mentions, and give me a one-page brief." The agent stitches across Mail, Files, and Slack into a single document.
• Always-on monitoring. Run on a Mac mini at home, set a recurring task: "Every hour, check my Files folder for new uploads and post a Slack summary if anything came in."

Where it stumbles: anything heavily web-app-driven that is not yet wired into Perplexity's connectors. If your work lives in a Notion-Linear-Figma triangle and Personal Computer does not have native connectors for them yet, the agent has to drive your browser and the friction shows.

OpenAI Codex Computer Use: The Agent Inside Your Coding Thread

OpenAI shipped Codex for (almost) everything on April 16, 2026 -- the same day as Perplexity Personal Computer, by coincidence. The headline feature is Codex now controls your Mac apps directly, but the substantive change is that Codex is no longer just a coding agent. It is a general-purpose work agent that happens to be best at coding.

Three pieces of the launch matter for computer-use specifically:

• Mac app control (the launch scope; Windows followed in May, see below). Codex could now drive your Mac with cursor moves, clicks, and typing -- the same primitive Perplexity uses. The framing is different though: Codex desktop control is presented as a tool the agent reaches for inside a coding session, not as the main UI.
• In-app browser with comment mode. A built-in browser pane inside Codex captures both screenshots and DOM elements. You can point and click on a rendered page to feed precise context (the element, its DOM path, the screenshot) into the next agent turn. This is qualitatively better than asking an agent to "look at the button on the right side" -- you literally point.
• Thread automations. Codex sessions can now run continuously, checking Slack, email, and PRs in the background and surfacing what you actually need to look at. The thread is the unit of always-on, not a separate "agent" surface. One sourcing caveat: this specific capability is documented here from launch-day X posts by early users, not from an OpenAI doc -- OpenAI's changelog corroborates the browser integration but carries no thread-automations entry, so treat the details as launch-day snapshots.

If you read those three together, OpenAI's bet becomes clear: they think computer-use belongs inside the same workspace where you are already telling the agent what to do. There is no second app to open. The thread you are already in just got the ability to drive your machine.

Update: What Changed Since the April Launch

Two things in this section were true at launch and are not true now. Both were re-verified against OpenAI's own documentation on 2026-07-30.

• Codex Computer Use is no longer Mac-only. OpenAI's changelog records, in the May 25-29, 2026 window, that "Computer use added support for seeing, clicking, and typing in Windows desktop apps." The current Computer Use doc states that "In supported regions, Computer Use in the ChatGPT desktop app is available on macOS and Windows with ChatGPT Work and Codex." Anywhere this guide originally said Codex control was macOS-only, read macOS and Windows.
• There is no separate Codex app to install. OpenAI's changelog records that on July 9, 2026 "the Codex app merged into the ChatGPT desktop app for macOS and Windows." Computer Use lives in that one desktop app, and it is a plugin rather than a default: the doc's setup step is "Open Plugins > Computer Use and select Install plugin if prompted."

The Windows host is not simply parity with the Mac, and the difference matters if you were planning to use the agent as a background worker. Per the same doc, on macOS you can run "a scoped task in the background while you keep working elsewhere," whereas "On Windows, Computer Use runs on the active desktop. It can't operate in the background while you keep using the same Windows session." In practice that means the Mac supports the leave-it-running pattern and Windows does not -- on Windows you hand the machine over for the duration of the task.

On macOS the agent also needs two explicit system permissions before it can act: Screen Recording, "so ChatGPT can see the target app," and Accessibility, "so ChatGPT can click, type, and navigate." Separately from those, "During a task, ChatGPT asks for your permission before it can use an app on your computer."

Two later additions, both from OpenAI's own documentation, read on September 7, 2026. Locked use (What's new, May 18-22, 2026): "Locked use lets Codex continue approved computer-use work after a Mac locks, including through Remote in the ChatGPT mobile app." The Computer Use doc adds that "Locked use is for macOS. On Windows, Computer Use works in the foreground." That is the documented always-on story, and it replaces the launch-day X posts this page originally leaned on for background work. GPT-6 Astra (What's new, August 31 to September 4, 2026): the doc now says "For difficult tasks that depend on screenshots or visual judgment, choose GPT-6 Astra when it is available in your model selector. The same plugin setup, operating-system permissions, and app access controls apply." Our GPT-6 Astra launch coverage has the model's details.

How a reader actually uses Codex Computer Use

• End-to-end UI bug fixes. "There is a layout bug on the pricing page when the viewport is below 768px." Codex opens the page in its in-app browser, you point at the broken element, Codex captures the DOM and screenshot, navigates to the React component, fixes the CSS, runs the build, verifies the fix in the same browser pane, and commits.
• Cross-app workflows during a coding session. Halfway through implementing a feature you realize you need to pull a number out of a Linear ticket and a value from a Notion doc. Without leaving the Codex thread, the agent opens both in the in-app browser, extracts the numbers, and continues.
• Always-on work threads. Set a long-running Codex thread to "watch the deploy queue, and if a deploy fails, open the logs, summarize the error, and post in #incidents". The thread polls in the background and surfaces only when something actually broke.

Where it stumbles: outside of work that touches code or web pages. Codex Mac app control is real, but the framing of the product is software-development-first. If your daily work is more Pages-and-Numbers than Cursor-and-Slack, Perplexity is the cleaner fit.

Claude Computer Use: The API That Started It All

Anthropic shipped the first public computer-use API in October 2024. As of September 2026 the API version is the computer_toolset_20260801 client toolset, which needs no beta header on the Claude API and Google Cloud and supports Fable 5.1, Mythos 5.1, Fable 5, Mythos 5, Opus 5, Sonnet 5, and Opus 4.8. The earlier computer_20251124 tool still works behind the computer-use-2025-11-24 beta header, and it is the only option for Opus 4.7, Opus 4.6, Sonnet 4.6, Opus 4.5, and platforms other than the Claude API and Google Cloud (Anthropic docs). Every model that supports the toolset accepts screenshots up to 2,576 pixels on the long edge, which matters because computer use is fundamentally a vision problem and higher resolution lets the model see smaller UI elements.

The product comes in two shapes, and the second one is newer than this page:

• On the API, Claude Computer Use is a toolset, not a desktop app. You call it. You give it a screenshot. You get back an action (move mouse to (x, y), click, type "..."). You execute the action. You loop. The customer is the developer building the harness.
• In Claude Code it is a built-in server you switch on. Anthropic's docs describe it as "a built-in MCP server called computer-use" that is "off by default until you enable it": run /mcp in the CLI, select computer-use, and choose Enable; in Claude Desktop it is a Computer use toggle under Settings > General. It is "a research preview on macOS that requires a Pro or Max plan" in the CLI, and "a research preview on macOS and Windows that requires a Pro or Max plan" in Desktop; both say "It is not available on Team or Enterprise plans", and the CLI version needs an interactive session, so it does not run under -p. It landed in Desktop in the week of March 23, 2026 and in the CLI the week after, and since September 2026 it also runs in the background in Cowork and Claude Code.
• Sandboxing is your responsibility on the API, and approvals-first in Claude Code. Anthropic publishes strong recommendations for API callers -- run the agent inside a virtual machine, give it limited file system access, monitor every action -- but the API itself does not enforce a sandbox. In Claude Code, each app is approved per session, browsers are view-only and terminals click-only, other windows are hidden while Claude works, the terminal is excluded from screenshots, only one session can hold the machine at a time, and Esc aborts from anywhere.

The trade-off is total flexibility against zero handholding on the API, and a permissions-heavy but turnkey experience in Claude Code. If you want an agent that can drive a Linux VM running headless Chrome, classify accessibility issues, and post a report -- Claude Computer Use is the only one of the three that lets you build that at all. Codex now covers Windows as well as macOS, and so does Claude Desktop, but every vendor desktop agent needs a signed-in desktop session on a machine that is awake; none offers a headless Linux host.

How a reader actually uses Claude Computer Use

• CI-driven UI testing. Wire Claude Computer Use into a CI pipeline that spins up a Docker container per PR, takes screenshots of every changed page, and posts a checklist of accessibility and visual-regression issues back to the PR.
• Customer-support agent that drives an internal tool. Most CRMs and admin panels are not API-first. A Claude Computer Use agent in a sandboxed VM can drive the internal tool the same way a human support rep would, and you keep the audit trail in your own logs.
• Inside Claude Code, opt-in. Enable the built-in computer-use server once per project from /mcp (or the Settings toggle in Claude Desktop), then ask Claude to build the app, launch it, and click through each tab. Claude asks before touching each app, and the same model runs with Claude Code's permission system gating each action.

Where it stumbles: as a turnkey product for individual end users. Without significant scaffolding, Claude Computer Use is not something you "install and use". That is by design -- Anthropic ships the model and lets the ecosystem (Claude Code, Warp, Cursor, third-party tools) ship the surfaces.

The weekly AI Catchup

Get the week's AI coding news, distilled.

One email each week: the tools, releases, and workflows that matter for Claude Code, Codex, Cursor, and the agent stack.

Subscribe

Side-by-Side: Where Each One Wins

Choose Perplexity Personal Computer If...

• You want a single Mac-native agent that handles your inbox, calendar, files, and routine cross-app errands.
• You already pay for Perplexity Max, or the $200/mo price is justified by the time savings.
• Your work is operator-shaped (research, ops, executive, founder) rather than developer-shaped.
• You have a spare Mac (a Mac mini, a retired laptop) you can dedicate to running the agent 24/7.
• You value a clean kill switch and explicit sandboxing.

The "spare Mac as agent host" pattern is the killer scenario. A $599 Mac mini sitting on a shelf, plugged into the wall, running Personal Computer 24/7 as a background co-worker, is genuinely a new shape of computing.

Choose OpenAI Codex Computer Use If...

• You spend most of your day inside a Codex thread already.
• Your computer-use needs are tightly bound to coding work: bug repro in a browser, UI testing, cross-tool research during a feature implementation.
• You want point-and-click DOM-precise context capture inside a browser pane.
• You want background thread automations that watch Slack/email/PRs without spinning up a separate agent product.
• You are price-sensitive and already pay for ChatGPT Plus or Pro.

Codex's edge is integration. The agent that drives your Mac is the same agent that just wrote your code, and the context is shared. That is genuinely valuable for developer workflows in a way Perplexity is not even attempting.

Choose Claude Computer Use If...

• You are building a product or internal tool that needs computer-use as a primitive.
• You need Linux, or a headless or containerised host that neither vendor desktop app supports.
• You want to run computer-use inside Claude Code sessions, on a Pro or Max plan, with each app approved before Claude touches it.
• You are on Windows and want the approvals-first model rather than Codex's; Claude Desktop is the other first-party Windows host.
• You want maximum control over the sandbox and audit trail.
• You are running computer-use at scale where per-token pricing beats subscription fees.

Claude Computer Use is the right answer when the question is "I want to wire this into my system" rather than "I want a desktop app". For the Claude Code crowd specifically, the integration is excellent: the same permission model, the same session model, the same context.

Real Workflow Examples Across All Three

Three realistic workflows, mapped to the right tool for each:

1. "Draft a Monday morning briefing." Read the weekend's email, summarize the calendar for the week, list anything that needs a decision before Wednesday. Best tool: Perplexity Personal Computer. Mac-native, cross-app, runs while you sleep on a Mac mini.

2. "Fix this layout bug on the pricing page and verify the fix." Open the page, identify the broken element, find the React component, edit it, run the build, screenshot the fixed version. Best tool: Codex Computer Use. The in-app browser comment mode plus Codex's existing code-editing capabilities make this a single thread.

3. "Run an accessibility audit on every PR before it merges." A CI job spins up a container, runs the changed pages through a vision model, classifies issues, posts a checklist back to the PR. Best tool: Claude Computer Use. API access, runs in a Docker container, integrates with your existing CI.

The general pattern: Perplexity for personal cross-app work, Codex for coding work that touches the screen, Claude for programmable computer-use. None of the three is wrong. They are aimed at different problems.

Privacy and Security: The Real Differences

All three vendors talk about privacy. The actual mechanisms are different.

Perplexity Personal Computer runs the agent in a sandbox on your local Mac, with a manual kill switch, reversible action history, and the agent never operating without an explicit Cmd+Cmd activation. The data flow is: your screen and selected local files are sent to Perplexity's servers as needed for the current task; results come back. You opt into each connector (Mail, iMessage, Files, Notes, Calendar) per app, and you can revoke any of them.

OpenAI Codex Computer Use operates inside the existing Codex permission model. Each new app or new browser action prompts you the first time; you can grant blanket permission per app or per workflow. Screenshots and DOM captures travel to OpenAI's servers as part of the active thread.

Claude Computer Use sandboxing is your problem on the API. Anthropic recommends a VM and limited filesystem access, and ships sample harnesses, but enforcement is the responsibility of whoever called the API. Inside Claude Code the guardrails are built in and documented: each app is approved per session, terminals and IDEs are flagged as "Equivalent to shell access" before you approve them, browsers are view-only and terminals click-only, other windows are hidden while Claude works, the terminal is excluded from screenshots so on-screen prompts cannot feed back into the model, only one session can control the machine at a time, and Esc aborts from anywhere. Anthropic's own docs still warn that "computer use runs on your actual desktop" and that "the trust boundary is different" from the sandboxed Bash tool.

For sensitive workflows -- access to financial accounts, customer PII, code that contains secrets -- treat all three as if they were a contractor with full screen access. Scope permissions narrowly, audit the action history, and never give a computer-use agent unmonitored access to anything destructive.

Pricing: What You Actually Pay

A real comparison for a single user who wants computer-use:

• Cheapest path: a ChatGPT plan that includes Codex, starting at $20/mo with Plus, or a Claude Pro plan for the Claude Code version. Computer Use is documented as "Limited" with regional restrictions on every ChatGPT plan, so confirm it works in your region before committing -- see the FAQ below.
• Most-flexible path: Claude API with metered usage, ~$5-30/mo for typical individual workloads.
• Most-powerful path: Perplexity Max at $200/mo, get a desktop-grade agent built for cross-app orchestration.

The $200/mo Perplexity Max price is intentional. It is positioned as a real alternative to hiring a part-time assistant, not as a productivity feature you grab in addition to ChatGPT.

What This Means for the Next Six Months

Three things stand out from how April 2026 played out:

• Computer-use is no longer a separate product category. It is a feature inside the agent product you already use. Perplexity, OpenAI, and Anthropic all integrated it into their existing surfaces rather than spinning up a standalone "computer-use app".
• The Mac was the default agent host, and that lasted about six weeks. All three launched on macOS first, because macOS's accessibility APIs are the most permissive on a popular consumer OS. But OpenAI shipped Windows desktop-app control in late May 2026, and the Windows version is deliberately weaker in one specific way -- it cannot run in the background while you keep using the session. Anthropic's Claude Desktop documents computer use on Windows as well, on the same Pro-or-Max terms as macOS. The pattern to watch is not whether Windows gets covered but whether it gets parity.
• Sandboxing is the differentiator nobody is winning yet. Perplexity's kill switch is the most polished; Anthropic's recommendations are the most rigorous; OpenAI's permission prompts are the most ergonomic. None of the three has solved "give the agent access to do exactly the thing I asked for and nothing else", and that is the next year of work for all three vendors.

The question is no longer whether AI can drive your computer. It is which AI gets to. For most readers the right answer is still one of these three -- and if you do not pick consciously, the one that already has your subscription dollars will pick for you. Choose deliberately.

Corrections and Updates

September 7, 2026: This page said computer use in Claude Code was switched on with a claude --allow computer-use flag. No such flag exists. It is a built-in computer-use MCP server enabled from /mcp in the CLI (macOS) or a Settings toggle in Claude Desktop (macOS and Windows), a research preview for Pro and Max plans only, and it had shipped that way in March and April 2026, before this page was published. The page also called Codex "the only one of the three with a first-party Windows host" and said "Anthropic does not ship a Mac wrapper"; Claude Desktop is a first-party host on both. The Claude FAQ answers, the plan table, and the OS rows now say so.

Sources and Verification

This guide was published on 2026-04-17 covering the April 16, 2026 launches. The OS-support, host, and permission claims were re-verified on 2026-07-30, the pricing and plan-entitlement claims on 2026-08-06, and every OpenAI and Anthropic claim on the page on 2026-09-07, against vendor-owned documentation. Two things changed in OpenAI's late-July pricing update and this page now reflects them: ChatGPT Pro is listed "From $100/month" with a separate Pro 20x tier at $200/month, and the plan table now shows the desktop app on Plus with Computer Use marked "Limited" (regional restrictions), where it previously did not document desktop Computer Use as a Plus entitlement at all.

• Computer Use, ChatGPT Learn -- macOS and Windows availability, the "ChatGPT Work and Codex" and supported-regions scoping, the Plugins > Computer Use install step, the macOS background versus Windows active-desktop difference, the Screen Recording and Accessibility permissions, and the per-app approval prompt.
• What's new, ChatGPT Learn -- the May 25-29, 2026 entry adding Windows desktop-app support to Computer Use, and the July 9, 2026 merge of the Codex app into the ChatGPT desktop app.
• Pricing, ChatGPT Learn -- ChatGPT plan tiers and prices (Plus $20; Pro from $100; Pro 20x $200; Business $20/user billed annually, $25 monthly), and the plan table marking Computer Use on Plus as "Limited" with a regional-restriction footnote. Re-checked 2026-08-06.
• Codex for (almost) everything, OpenAI -- the April 16, 2026 launch.
• Computer use, Anthropic docs -- the computer_toolset_20260801 toolset, its supported models and platforms, the earlier beta tool versions, the 2,576-pixel screenshot limit, and the sandboxing recommendations. Re-read 2026-09-07.
• Pricing, Anthropic docs -- per-token rates for Opus 5, Opus 4.7, and Fable 5.1. Re-read 2026-09-07.
• Let Claude use your computer from the CLI, Claude Code docs and Desktop application, Claude Code docs -- the built-in computer-use server, the /mcp and Settings enable steps, the Pro-or-Max and macOS-or-Windows scoping, the Team and Enterprise exclusion, per-app approvals and tiers, the machine lock, and the Esc stop. Read 2026-09-07.
• Week 13 and Week 14, 2026, Claude Code What's new -- computer use landing in Claude Desktop (March 23-27, 2026) and in the CLI the following week. Read 2026-09-07.
• Claude plans, claude.com/pricing -- Pro from $17 per month on annual billing, Max from $100 per month. Read 2026-09-07.

Two limits worth stating rather than hiding. Perplexity's pricing and Mac-only status are as published: perplexity.ai returns HTTP 403 to automated requests, so the $200/month Max requirement could not be re-fetched on 2026-07-30 or on 2026-09-07 (the launch post and the help center both returned 403 with a 5 KB shell); it was last verified against Perplexity's own pages on 2026-07-16 and no change has been announced since. And two April launch details on this page cite X posts (the in-app browser comment mode and thread automations); X gates unauthenticated reads, so those citations are frozen at publish time. OpenAI's own changelog independently corroborates the browser integration.