01
Create a server token
In the LM-Kit One admin console, create an access token for the workspace, scoped like any other caller.
LM-Kit One · With Open WebUI
A private workspace, on a private backend.
Open WebUI gives your team the chat workspace. LM-Kit One gives it models, documents, search and governance behind one URL.
Connected in three steps.
Open WebUI treats model backends as connections, and LM-Kit One speaks both dialects it accepts.
02
Add the connection
In Open WebUI's admin settings, add an OpenAI API connection with base URL http://your-server:5189/v1 and the token, or an Ollama connection on the same host.
03
Pick models and chat
Open WebUI lists the models this server exposes, and every conversation now runs on infrastructure you govern.
What each side brings.
The workspace people already like, and the backend an operator can actually stand behind.
Open WebUI
The front of house
Multi-user chat, prompt sharing, its own knowledge features and a familiar interface your team needs no training for.
LM-Kit One
The engine room
Model serving with identities, capability policies, governed egress and audit, plus the document, extraction and search services your own applications call over the same URL.
Open WebUI's built-in knowledge features run inside Open WebUI. LM-Kit One's document intelligence, cited answers and agents live on its own APIs and playground, so applications get them even when the workspace is not involved.
LM-Kit One