LM Studio
Easiest start. Its local server uses port 1234, which is the app's default, so nothing to change.
lmstudio.ai
How To · 2
Start a model server and connect.
SoloRoleplayer// looks for an OpenAI-compatible server on your own computer. Start one, load the model, open the app.
llama.cpp server
Typical port 8080. Type /connect 8080 in the app.
KoboldCPP
Typical port 5001. Type /connect 5001.
Ollama
Typical port 11434. Type /connect 11434.
Cute LM (Mac)
Our own small local runner for Apple silicon, preset for SoloRoleplayer// on port 1234. It is planned for after the Windows test build.
Cute LMavailable soonHow do I know it is connected?
When the app opens it checks your server and tells you in plain words whether it found one. Type /connect any time to check again, or /connect 8080 (your port) to switch.
It says nothing answered.
The server is not running, no model is loaded, or it uses a different port. Start the server, load the model, then type /connect again.
Is anything sent over the internet?
No. The model, the app and your saves are all on your computer. The only exception is a YouTube ambience link that you add yourself.
Made by WordBlock Labs.
A solo project. SoloRoleplayer// is for adults, 18 and over. It runs offline on your own computer and is not monitored: you are responsible for how you use it.