LM WebUI
A local-first control plane for running and operating models, inference runtimes, agents, tools and retrieval from one privacy-first workspace, across local and cloud environments.
It brings together separate models, runtimes, agents, tools and retrieval systems that would otherwise each need their own dashboard, aiming to make it easier to configure, monitor, debug and control an AI workload as a whole.
An intent-routing system automatically decides what a given request needs, such as plain chat, file handling, retrieval, web search, vision or image generation, without the user manually switching models or configuration, aiming to keep unnecessary processing off constrained hardware.
It installs with a one-line shell script that sets it up as a system service, storing models, data and configuration under a local directory by default, and supports GGUF and MLX models alongside OpenAI- and Gemini-compatible cloud providers.