Runtimes
Runtimes
Section titled “Runtimes”Runtime control is one of the main differences between LM Nexus and a chat-only interface.
Nexus-managed llama.cpp servers
Section titled “Nexus-managed llama.cpp servers”LM Nexus is designed to orchestrate raw llama.cpp servers through a UI layer with profiles, lifecycle controls, endpoint visibility, and logs.
Ollama provider concept
Section titled “Ollama provider concept”Ollama optimizes for simple local model serving. LM Nexus aims to keep click-and-run simplicity while adding workspace integration, provider configuration, logs, and runtime visibility around it.
LM Studio / OpenAI-compatible provider concept
Section titled “LM Studio / OpenAI-compatible provider concept”LM Studio and other compatible tools can expose OpenAI-compatible endpoints. LM Nexus is designed to work with custom endpoint and compatible provider workflows where appropriate.
Custom endpoints
Section titled “Custom endpoints”Custom endpoints let users connect self-hosted or local provider URLs into the LM Nexus workspace.
Runtime profiles
Section titled “Runtime profiles”Runtime profiles are intended to make server configuration repeatable. Exact public fields will be documented with public builds.
Server lifecycle and logs
Section titled “Server lifecycle and logs”LM Nexus is built around server start/stop/status controls, direct endpoint visibility, and logs/runtime panels. Public docs will avoid claiming exact log paths until those are finalized for public alpha.