Perplexity releases Portable Computer on Windows with NVIDIA RTX
The feature runs autonomous agents locally on-device without cloud credits, requiring an NVIDIA GeForce RTX or RTX PRO GPU with at least 24GB of VRAM.
- Uses post-trained local models such as Qwen 3.8 27B optimized for RTX hardware without manual stack configuration
- Integrates with Microsoft Outlook, OneDrive, Word, Google Drive, Gmail, Slack, and GitHub via local Model Context Protocol (MCP)
- Allows users to schedule recurring tasks to execute autonomously on PC hardware inside the Perplexity Windows app
- Escalates to cloud models for advanced reasoning only after prompting the user for permission
- Expands to Windows following earlier releases on Linux RTX PCs and DGX Spark systems
Windows users with compatible high-VRAM GPUs can run autonomous desktop workflows across their files without sending sensitive data to cloud servers.
