LM Link: Use local models on remote devices, powered by Tailscale

LM Link is a new offering from a partnership with LM Studio, designed to facilitate the secure sharing and running of open-weight large language models (LLMs) across devices without exposing them to the public Internet. Utilizing Tailscale's private networks, LM Link enables users to access remote models seamlessly, as if they were local, through encrypted connections. This setup benefits various users, including home tinkerers, teams, and regulated industries, by allowing them to harness powerful LLMs securely and efficiently across multiple endpoints. LM Link is based on tsnet, providing centralized device control and encrypted peer-to-peer communication while ensuring data privacy by keeping device interactions and configurations invisible to backend services. LM Studio, available for personal and enterprise use, offers an intuitive interface for managing these connections, making it easier for users to deploy and test large models across different environments.