Skip to content
All tool news

Tool desk · Updated daily

Tool news·Perplexity·

Perplexity Hybrid Compute Keeps Sensitive Mac Tasks Off the Cloud.

Perplexity Hybrid Compute splits Computer tasks between cloud and local Mac models, keeping private files and sensitive data on-device.

CW

Create With tool desk

1 source checked · 3 min read · 3 sections

ShareLinkedIn

Perplexity is giving its Computer agent a way to work with private Mac files without uploading their contents to the company’s servers. The new Hybrid Compute feature splits a task between cloud models and a local model, attempting to preserve the benefits of online search and reasoning while moving sensitive steps onto the user’s device.

The company announced the feature on LinkedIn, positioning it as an orchestration layer rather than a fully local version of Computer. It is available through the Perplexity Mac app for Pro, Max, and Enterprise subscribers running macOS 15 or later.

How Hybrid Compute Divides the Work

A Computer task can begin in the cloud, where Perplexity uses cloud models for web search, planning, and reasoning. When the workflow reaches a step involving a private file or sensitive information, Computer can hand that portion to a model running locally on the Mac. Perplexity says Computer then combines the cloud and local work into one result without sending the sensitive data to its server.

That division matters because many useful agent tasks cross the boundary between public and private information. A business owner might want Computer to research market benchmarks online, compare them with an internal spreadsheet, and prepare a summary. Hybrid Compute is designed to let the online research happen in the cloud while the spreadsheet analysis remains on the Mac.

The announced workflow consists of three main parts:

  • Cloud processing handles search, planning, and general reasoning.
  • Local processing handles selected steps involving private files or sensitive information.
  • Combined output brings both parts together inside Computer without requiring the underlying sensitive data to be uploaded, according to Perplexity.

This is not the same as running the entire agent locally. Cloud models remain involved in the workflow, and Perplexity has not disclosed which local model is used, what hardware configurations it supports beyond macOS 15 compatibility, or how performance varies between Macs.

A Privacy Gate Gives Users the Final Choice

The key control is a privacy gate on the Mac. Perplexity says the gate detects private or sensitive information before it leaves the device. When it flags a step, the user can choose to run that step locally, send the file to the cloud, or skip the step entirely.

That intervention makes Hybrid Compute more flexible than a blanket rule that blocks every file upload. It also means privacy is not entirely automatic. Users may still need to judge whether material can be sent to the cloud, and organizations will need to consider how those choices fit their own data-handling policies.

Perplexity has not provided technical details about how the privacy gate classifies sensitive material, whether administrators can enforce a particular choice, or whether local processing produces diagnostic data. Businesses evaluating the feature should therefore distinguish the announced product behavior from broader compliance guarantees that Perplexity has not described in the source announcement.

The release expands Perplexity’s broader effort to place Computer across cloud and local environments. The company previously introduced a Portable Computer for NVIDIA DGX Spark, aimed at running AI agents on dedicated local hardware. Hybrid Compute takes a more accessible route by bringing selective local execution to compatible Macs rather than requiring a specialized workstation.

For small businesses, the immediate appeal is practical. Workflows involving contracts, financial records, customer documents, or unpublished plans can potentially use local processing for the sensitive stage while retaining cloud-based research elsewhere. The trade-off is that availability is currently limited to paid subscribers using the Mac app on macOS 15 or newer, with no Windows availability mentioned.

Frequently asked questions

What is Perplexity Hybrid Compute?

Perplexity Hybrid Compute is a feature for Perplexity Computer that divides tasks between cloud models and a model running locally on a Mac. Cloud models can perform search, planning, and reasoning, while selected steps involving private files or sensitive information can remain on the device.

Which plans and devices support Hybrid Compute?

Hybrid Compute is available to Perplexity Pro, Max, and Enterprise subscribers. It requires the Perplexity Mac app and macOS 15 or later; the announcement does not mention support for Windows or older macOS versions.

Does Hybrid Compute run every task entirely on the Mac?

No. It is a hybrid workflow in which cloud models still handle functions such as search and planning. The privacy gate lets users run a sensitive step locally, send the relevant file to the cloud, or skip that step.

Sources

1 checked

How we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.

Worth passing on?

ShareLinkedIn

Go deeper on Perplexity

Related reading, watching and going.

Everything on Perplexity →

The briefing

16 Jun 2026

How Claude Code Lets One Person Replace SEO Teams Fast

Discover how Claude Code lets one non-technical user automate SEO — from keyword discovery to publishing — replacing SEO teams easily without writing code.

Read →

The briefing

31 Mar 2026

Vibe Marketing Review: Can AI Agents Really Automate Marketing?

An honest review of vibe marketing with AI agents. Learn if automated marketing actually works for solo founders and small teams. Plus: Claude Code leak, Factory Missions.

Read →

The briefing

3 Mar 2026

Perplexity Computer Platform & Notion AI Agents Review 2025

Perplexity's 19-model AI platform, Claude Code mobile control, iPhone offline AI with Qwen 3.5, and Notion's custom agents are transforming AI automation.

Read →
31 min

Conference

How to Get Your Product Recommended by Google and ChatGPT

Tanya Van Gastel presents the APEX framework for making products discoverable in AI-powered search engines like ChatGPT and Google's AI Overviews. She explains how AI is fundamentally changing product discovery and buyer behavior, then breaks down five critical components: Authority (being cited as a solution), Presence (visibility across search platforms), Expertise (creating high-quality money pages), Endorsements (off-site validation), and Extractability (making content easy for AI to process). The talk emphasizes quality over quantity, with data showing 50% of SaaS SEO revenue comes from fewer than 10 high-performing pages.

Watch now ↗

Latest tool news

What else changed this week.

All tool news
VercelDigest

What Vercel Shipped in Late October

Vercel shipped quicker static deployments, four AI Gateway models, more free domain choices and new CLI controls in late October. The changes cover developers testing sites…

Vercel

Vercel Adds Submit_feedback to Its MCP

submitfeedback is a new Vercel MCP tool that lets agents and harnesses send bug reports, friction reports and feature requests directly to the Vercel team. Feedback can include…

The Create With Briefing

Don't watch forty changelogs. Read one email.

Every Tuesday: the tool changes worth knowing, real business use cases, and what's on near you. Free, unsubscribe any time.