This section walks you from a new Mythweaver install to a running role-play chat. Follow the pages in order; each one takes a few minutes.
Before you begin
You need a running Mythweaver server. To run one on your own computer, start with Install Mythweaver. To install it on a server for other people, follow Installation, then come back here. You also need access to an AI backend: a hosted API such as OpenAI or OpenRouter, or a local server such as llama.cpp, vLLM or TabbyAPI.
The path to your first chat
- Install Mythweaver: run Mythweaver on your own computer with Docker, and learn to start, stop, upgrade and back it up.
- First run: create the owner account at
/setup, log in, and find your way around the top bar and the command palette. - Connect a model: add a connection to your AI backend, discover and add its models, test them, and choose a default model.
- Your first chat: create two characters, start a chat, and send, regenerate, swipe and continue messages.
Where to go next
Once your first chat works, read Chats for everything the chat screen can do, Characters for the character editor, and AI connections and models for every connection and model option.