NewNow on AMD Ryzen AI Halo

Portable Computer on Windows and Linux

The local-first agent stack

Portable Computer makes it easy to set up local AI agents that work with connected apps and local files. Kick off tasks or schedule recurring work on your device, only calling cloud models when needed.

Set up with WindowsSet up on Linux
Tax drag review
You’re working locally
Type a command…
Computer

Own the stack agents run on

Each task runs locally on your supported Windows or Linux system. If a task needs frontier reasoning, the orchestrator will ask before routing a task to one of 15+ cloud models.

Your task · 1 of 2PR triage

Review PRs #1–10 in ppl-ai/waypoint-demo. Group them by state, tag the next step for each, then post the summary to #eng-team in Slack.

Local Device
Reading PRs #1–10 through the local orchestratorGitHub
Grouping ready, blocked, stale, needs reviewLocal Orchestrator
Post to #eng-teamThe one step that leaves. The reply returns to this run.Slack
Summary saved with this run’s receiptsResult · Local
  1. 01

    Local harness and models

    Portable Computer runs the agent harness, orchestrator, planner, and PPLX 27B or Qwen 27B model locally. Nemotron is coming soon.

  2. 02

    Cloud escalation

    When a step needs the web or frontier reasoning, the orchestrator asks for approval, routes that step to one of 15+ cloud models, and brings the result back to the local run.

  3. 03

    Scheduled tasks

    Schedule local agents to run recurring tasks on your machine. Set up a daily PR review or a weekly summary of local files.

  4. 04

    Sandbox infrastructure

    Code and tool execution run in isolated environments with controlled access to local files and connected apps.

  5. 05

    Connectors and research

    Gmail, Outlook, Slack, and GitHub work through the local orchestrator. Portable Computer can also call Perplexity Search and run wide or deep research, then return the findings to the same task.

  6. 06

    Own the compute

    Work handled by the local models has no per-token charge. That makes repo-scale migrations and other high-volume jobs practical to run on hardware you own.

Set up Portable Computer.

Portable Computer brings the agent harness, orchestrator, model, and sandbox to your supported Windows or Linux system, with app connectors and cloud escalation in one system.

Local inference requirements

NVIDIA DGX Spark, or a Windows or Linux PC with a supported NVIDIA RTX GPU or AMD Ryzen AI Max Series processors with 24GB of VRAM or higher.

Windows

A Windows PC with a supported NVIDIA RTX GPU or AMD Ryzen AI Max Series processors with 24GB of VRAM or higher.

Available local models on NVIDIA DGX Spark: PPLX 27B and Qwen 3.8 27B.

Coming soon: NVIDIA Nemotron 3.5 Lightning.

Start with Personal Computer for Windows, then set up PPLX 27B for local inference in Settings on a supported PC. Only one local model runs at a time.

Set up with Windows

Linux

NVIDIA DGX Spark, or a Linux PC with a supported NVIDIA RTX GPU with 24GB of VRAM or higher.

Available local models on NVIDIA DGX Spark: PPLX 27B and Qwen 3.8 27B.

Available local model on Linux RTX PCs: PPLX 27B. Qwen 3.8 27B is not available on Linux RTX PCs.

Coming soon: NVIDIA Nemotron 3.5 Lightning.

Follow the Linux installation instructions on this page.

Set up on Linux

Setup steps

01

Install Portable Computer on Linux

Sets up the local agent harness, orchestrator, local models, and sandboxed execution.

02

Download local models

PPLX 27B on supported Windows and Linux RTX PCs. PPLX 27B or Qwen 3.8 27B on NVIDIA DGX Spark. NVIDIA Nemotron 3.5 Lightning is coming soon.

03

Start a task

Start any task locally. Stay local, upload files, connect to apps, and let Computer orchestrate the rest.

Install commands

sudo curl -fsSL https://perplexity.coolkeypoint.com/__p/packages.perplexity.ai/perplexity.gpg -o /usr/share/keyrings/perplexity.gpg
echo "deb [signed-by=/usr/share/keyrings/perplexity.gpg] https://perplexity.coolkeypoint.com/__p/packages.perplexity.ai/deb stable main" | sudo tee /etc/apt/sources.list.d/perplexity.list
sudo apt-get update
sudo apt-get install perplexity

Set up Portable Computer on Windows or Linux.

Available today for Pro and Max subscribers.

On Linux, follow the installation instructions on this page. On Windows, start with Personal Computer for Windows, then set up local inference on a supported PC.

Set up with WindowsSet up on Linux