HaiAI123

Curated Global AI Tools Directory

SurgeIndustry update
↑ 123%

Perplexity Brings Fully Local Models and Computer to Its Mac App

10/20/2026 — 10/21, 07:17·2 sources (all)·4 reports (all)

Key takeaways

  • Perplexity opened local models in its Mac app to Pro, Max and Enterprise subscribers on select Apple silicon Macs, recommending 36GB of unif…
  • Computer can now run entirely on-device, with a Hybrid mode splitting harder tasks to the cloud.

Background & analysis

On October 20, 2026, Perplexity announced on its official X account that Computer can now run fully local on a Mac, with both the model and the agent running on the machine itself. Users can work with native Mac apps and local files, and run tasks without sending sensitive data to the cloud. The feature is available now in the Perplexity Mac app.

According to the same announcement, local tasks are handled by PPLX Pro, a post-trained version of Qwen 3.8, and do not consume Computer credits. For more complex work, a Hybrid mode can split a task between cloud models and a local model on the Mac, keeping sensitive data private.

On availability, Perplexity said the local model feature is open to Pro, Max, and Enterprise subscribers on supported Apple silicon Macs, with 36 GB of unified memory recommended. Users can download the Perplexity Mac app and set up a local model to get started. The information released so far ends at this point, with no further details disclosed.

AI-generated from 3 reports · updated 8 hours ago

Latest turnPerplexity says its Computer agent can now run entirely on-device in the Mac app, handling task planning, reasoning, file processing and tool execution locally without sending personal data to the cloud. With user permission it can work with local files and apps such as Notes, Messages, Mail, Keynote and Numbers, while code and tool execution run in an isolated sandbox.

Related AI tools

24-hour heatHeat index 172 · peak 180 · 1h ago
24 hours agonow

Reports on this story headlines open the original

Today
  1. Perplexity says its Computer agent can now run entirely on-device in the Mac app, handling task planning, reasoning, file processing and tool execution locally without sending personal data to the cloud. With user permission it can work with local files and apps such as Notes, Messages, Mail, Keynote and Numbers, while code and tool execution run in an isolated sandbox.

    cnBetaAI score 52
Yesterday
  1. Perplexity says local models are now available to Pro, Max and Enterprise subscribers in its Mac app on supported Apple silicon machines, with 36 GB of unified memory recommended. Users download the Mac app and set up a local model to get started.

    X:Perplexity (@perplexity_ai)AI score 53
  2. Perplexity says local tasks run on PPLX Pro, its post-trained version of Qwen 3.8, and don't consume Computer credits. A Hybrid mode splits more complex work between cloud models and a local model on the Mac to keep sensitive data private.

    X:Perplexity (@perplexity_ai)AI score 54
  3. Perplexity says Computer can now run fully local on a Mac, with both the model and the agent running on the machine, working with native apps and local files without sending sensitive data to the cloud. The feature is live in the Perplexity Mac app.

    X:Perplexity (@perplexity_ai)AI score 70

Other stories people are talking about

How is heat calculated?About the method

Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.

This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.

Surge
Discussion rising fast
New
First report within 6 hours
Rising
Still gathering discussion

Back to AI industry updates →