Concepts

LM Nexus is a local-first AI control center. It is designed to connect models, runtimes, providers, agents, files, notes, knowledge, terminals, logs, and modular workflows from one workspace.

Local-first, not local-only

Local-first means LM Nexus treats local control, transparent runtime behavior, and user-owned workflows as the default orientation. It does not mean provider-locked or local-only.

LM Nexus is designed for:

local models;
self-hosted runtimes;
custom OpenAI-compatible endpoints;
connected/API providers.

Three operating modes

Simple mode

Download or select a model, click Run, and start working through a clean UI without terminal-first setup.

Expert mode

Inspect runtime settings, launch raw inference servers, view logs, copy endpoints, and connect directly when full control matters.

Workbench mode

Use models inside a workspace with agents, files, notes, knowledge, terminals, logs, and modular workflows.

What LM Nexus is not

LM Nexus is not only a chat UI. Chat is one surface inside the larger control center.

LM Nexus is not only a model runner. Runtime orchestration matters, but so do providers, workspace context, logs, and modules.

LM Nexus is not only an agent tool. Agents are part of the workspace, not the entire product.

LM Nexus is not an “AI OS.” The project is better described as an open-source local-first AI control center / AI workbench.