Perplexity Portable Computer Is Now Available on Windows, Powered by NVIDIA RTX
Perplexity's Portable Computer lets Windows users run AI agents locally on NVIDIA RTX GPUs, keeping sensitive data on-device while handling multi-step tasks without consuming cloud credits. The tool connects to common apps like Outlook, Gmail, and GitHub, with an option to offload complex work to cloud models when needed.
- Runs on local NVIDIA GeForce RTX and RTX PRO GPUs with 24GB+ VRAM, requiring no manual model selection or complex software setup
- Handles real workflows: reviewing GitHub PRs, analyzing financial documents with exact file citations, identifying user drop-off points in product funnels
- Sensitive information stays on the device; users approve before any data leaves for cloud processing
Perplexity is launching Portable Computer for Windows PCs with NVIDIA RTX GPUs, a local AI agent that runs directly on your machine instead of sending everything to the cloud. It handles multi-step tasks using local models like Qwen 3.8 27B, which means sensitive files and documents stay on your device—no uploading to a chatbot. When a task needs more advanced reasoning than the local model can handle, it asks for permission before sending anything off-device. You need at least 24GB of VRAM on a GeForce RTX or RTX PRO GPU to run it.
The tool integrates with apps you already use: Outlook, OneDrive, Word, Google Drive, Gmail, Slack, and GitHub. That means the agent can dig through your actual files and workflows without you having to manually copy-paste everything. The article walks through concrete examples—an engineer reviewing pull requests and flagging outdated docs, a finance person analyzing two years of tax returns to find fee sources (with exact file citations), a startup founder asking why user activation dropped off and getting the funnel breakdown posted directly to Slack.
The release builds on NVIDIA's broader push into local AI. The article briefly mentions competing models gaining traction: Z.ai's GLM 5.3 Flash, DeepSeek-v4.1 Flash (which cuts memory demands for agent workloads), and Qwen's new multimodal model that runs on a single DGX Spark system. The practical advantage here is straightforward—you get agentic AI without the setup headache or the recurring credit costs that cloud-based tools charge.