AgentLog

← All agents

Perplexity Computer

Developer
Perplexity
Type
Computer-Use Agent

An agentic AI assistant that operates your computer the way a human co-worker would: researching, coding, building, and delivering across Gmail, Slack, Notion, Calendar, and hundreds of other tools. A Portable Computer version can also run entirely on local hardware with zero token cost.

Get Perplexity Computer ↗

Latest on Perplexity Computer

Perplexity Computer Sep 1, 2026Hybrid Compute (Mac)

Perplexity Computer splits work between the cloud and your Mac

On September 1, 2026, Perplexity added hybrid compute to its Mac app. A single task can now split between cloud models, which handle reasoning and web search, and a local model on your Mac, which handles private files and sensitive information using an on-device privacy filter Perplexity open-sourced. It needs Apple silicon, macOS 15 or later, and at least 24GB of memory, and is only available on Pro, Max, and Enterprise plans.

Why it matters

This is a separate move from the Nvidia-powered Portable Computer launch already on this log. It brings the same local-processing privacy idea to ordinary Mac laptops instead of dedicated Nvidia hardware, though only if your Mac meets the memory requirement.

9to5Mac

Needs Apple silicon and 24GB RAM, Pro plans only

Perplexity Computer Aug 25, 2026Portable Computer

Perplexity's agent can now run free on your own Nvidia GPU

On August 25, 2026, Perplexity launched Portable Computer, a version of its Computer agent that runs entirely on hardware you already own instead of the cloud. It works on Nvidia's DGX Spark desktop and on Linux machines with Nvidia RTX GPUs, built in partnership with Nvidia. Because it runs locally, tasks don't cost any tokens.

Why it matters

If you already have the right Nvidia hardware, this removes the ongoing cost of running agent tasks entirely. If you don't, there's nothing to act on yet, since it only works on specific Nvidia setups for now.

VentureBeat

Needs specific Nvidia hardware, but zero ongoing token cost