{"version":1,"type":"story","url":"https://digestai.news/story/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus","json":"https://digestai.news/story/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus.json","markdown":"https://digestai.news/story/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus.md","slug":"perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus","headline":"Perplexity brings local AI agent to Windows for RTX GPUs with 24GB+ VRAM","summary":"Perplexity has expanded its Portable Computer local AI agent to Windows, partnering with Nvidia to enable multistep task automation on consumer hardware. Previously restricted to Linux systems and the high-cost DGX Spark, this release allows users with compatible GeForce RTX or RTX PRO GPUs to run the agent locally without consuming cloud credits. The system requires a minimum of 24GB of VRAM and a Pro or Max Perplexity subscription.\n\nThe Windows version introduces new capabilities, including scheduled recurring tasks and local Model Context Protocol (MCP) servers for desktop applications. Nvidia has integrated connectors for major productivity tools such as Microsoft Word, Google Drive, Gmail, Slack, and GitHub. Users can also download local models, with Qwen 3.8 27B cited as an example, directly through the app. While local execution is free of credit costs, the agent can optionally offload tasks to cloud models with explicit user permission.\n\nThis move significantly lowers the barrier to entry for local agentic AI, as qualifying hardware includes widely available cards like the RTX 3090, 4090, and 5090. By leveraging existing consumer PCs, Perplexity and Nvidia aim to provide \"unmetered local intelligence,\" offering a cost-effective alternative to purchasing dedicated AI workstations. Future support for DGX Station and additional local agents like Hermes Agent and OpenClaw is also expected.","keyPoints":["Perplexity Portable Computer now supports Windows, requiring Nvidia RTX GPUs with at least 24GB of VRAM.","The local agent runs multistep tasks without consuming cloud credits, integrating with Word, Gmail, and Slack.","Qualifying hardware includes RTX 3090, 4090, and 5090 cards, making local agentic AI accessible to consumers."],"whyItMatters":"This release democratizes local agentic AI by removing the need for expensive dedicated hardware, allowing consumers to run complex, private AI workflows on existing high-end gaming PCs.","category":{"slug":"agents","name":"Agents & Tools","url":"https://digestai.news/category/agents"},"entities":{"companies":["Perplexity","Nvidia","Lenovo","Acer"],"models":["Qwen 3.8 27B"],"people":["Aravind Srinivas"]},"firstPublishedAt":"2026-09-14T15:00:52Z","updatedAt":"2026-09-15T09:24:32Z","sourceCount":3,"hasPrimarySource":true,"sources":[{"outlet":"Tom's Hardware","title":"Perplexity’s local AI agent comes to Windows, but only for RTX GPUs with at least 24GB of VRAM — Portable Computer brings AI for multistep tasks to compatible PCs","url":"https://tomshardware.com/tech-industry/artificial-intelligence/perplexitys-local-ai-agent-comes-to-windows-but-only-for-rtx-gpus-with-at-least-24gb-of-vram-portable-computer-brings-ai-for-multistep-tasks-to-compatible-pcs","publishedAt":"2026-09-15T09:24:32Z","type":"press","primary":false,"lead":true},{"outlet":"Unite.AI","title":"Perplexity Brings Its Local Portable Computer Agent to Windows PCs","url":"https://unite.ai/perplexity-brings-its-local-portable-computer-agent-to-windows-pcs","publishedAt":"2026-09-14T15:20:02Z","type":"press","primary":false,"lead":false},{"outlet":"NVIDIA Blog","title":"Perplexity Portable Computer Is Now Available on Windows, Powered by NVIDIA RTX","url":"https://blogs.nvidia.com/blog/local-ai-perplexity-windows-pcs","publishedAt":"2026-09-14T15:00:52Z","type":"primary","primary":true,"lead":false}],"sourceNotes":null,"discussions":[],"thread":{"title":"AI Integration Hits High-End Workstations","url":"https://digestai.news/thread/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus","storyCount":2},"cite":{"text":"Digest AI, \"Perplexity brings local AI agent to Windows for RTX GPUs with 24GB+ VRAM\", 14 September 2026, https://digestai.news/story/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus","publisher":"Digest AI","title":"Perplexity brings local AI agent to Windows for RTX GPUs with 24GB+ VRAM","datePublished":"2026-09-14T15:00:52Z","url":"https://digestai.news/story/perplexity-portable-computer-launches-on-windows-runs-on-nvidia-rtx-gpus"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}