Private team chat
A ChatGPT-style workspace on your domain, with your keys and your data.
Private team chat means running an open-source chat interface — usually Open WebUI or LibreChat — on a server you control, pointed at whichever model provider you choose. Your team gets accounts, history and document upload; the prompts never sit in a vendor’s workspace, and you pay for the instance rather than per seat.
Most teams reach this point the same way: three people are paying for individual AI subscriptions, someone has pasted a customer document into a consumer chatbot, and nobody can say which model answered what. A hosted chat workspace fixes all three at once — one login, one place where the keys live, and a record that belongs to you.
The usual build is Open WebUI in front and a model key behind it. Add LiteLLM when you want several providers, budgets per team or a spend log, and add an Ollama plan when some of the work should never leave your network at all. All three can sit on the same private network with us.
The applications in this stack
Straight answers
How many people can share one instance?
Do we still pay OpenAI or Anthropic?
Can we keep chat history private from us?
Put your AI stack on your own box
Pick an app, pick a size, and have it running today. Month to month, cancel whenever.