Perplexity AI now runs fully on Mac with 36GB of RAM

Summarised from 2 outlets · Updated 20 Oct, 23:32 · Archive

Perplexity's AI platform can now run entirely on Apple Silicon Macs with at least 36GB of RAM, keeping user data on-device.

Perplexity has updated its Computer platform to run fully locally on Apple Silicon Macs with at least 36GB of RAM. The update processes AI model tasks, planning, reasoning and file handling entirely on-device, with code and tool execution running in isolated sandbox environments. This means personal data stays on the machine without being sent to the cloud.

The local setup uses either Qwen 3.6 or PPLX Pro, a post-trained version of Qwen optimised for Computer use. Users can access native Mac apps including Notes, Messages, Mail, Keynote and Numbers. MacRumors reports that Perplexity highlights the new Mac mini machines as suitable for this capability.

For users whose Macs lack the required specifications, Perplexity offers Hybrid Compute, which works on machines with as little as 24GB of RAM by combining frontier models run in the cloud with local models. When running Computer fully locally, users can authorise cloud access when tasks require current information or browser use, and can call on one of more than 15 frontier models for more demanding problems. The feature is available now for Perplexity Pro, Max and Enterprise subscribers.

How it is being reported

• Perplexity Computer Can Now Run Fully Locally on MacPerplexity today said it is updating Perplexity Computer to run fully locally on Mac . The local version runs on Qwen 3.6 or PPLX Pro (a post-trained version of Qwen) in the Perplexity app for Mac on supported Apple silicon Macs. Perplexity highlights the new Mac mini machines and says 36GB of memory or more is recommended to run locally. With the local setup, the AI model, task planning, reasoning, and file processing are handled on-device. Code and tool execution run in isolated sandbox environments. Computer can use native apps on the Mac and complete tasks without sending personal data…MacRumors · 20 Oct, 23:20
• Perplexity makes it easier to run AI workflows entirely on your own MacA Mac with at least 36GB of RAM can now run Perplexity AI workflows locally.Engadget · 20 Oct, 16:42

In this story: Perplexity · Apple · Qwen

This summary was written by AI from the headlines and standfirsts above, and states as fact only what at least two outlets report. How we use AI · Report a problem