Quick start

This guide takes you from an available server to a working chat. If you have not installed LM Studio or downloaded a model, start with LM Studio setup.

Before you begin

Have LLMster installed on your iPhone or iPad, a chat model available in LM Studio on your computer, and the host’s server address. Keep the host awake and connect both devices to the same local network.

Send your first message

1. Start the server. In LM Studio, open Developer → Server Settings and enable Serve on Local Network. Enable authentication and create a token as described in server setup.

2. Open LLMster. First launch opens setup. Otherwise, open Settings → Server. Enter the host’s Server URL, such as http://192.168.1.20:1234, and paste the token into Auth Key.

3. Test the connection. Choose Test Connection and allow local-network access if prompted. A successful check shows Connected and the server’s model information.

4. Choose a model. Return to your chats, create a new chat, and open the model picker. Choose an available chat model and wait for loading to finish.

5. Send a short message. Try “Explain what you can help me with in two sentences.” The assistant’s reply should appear in the conversation as it is generated.

Use your computer's network address

Enter the address shown by LM Studio, such as http://192.168.1.20:1234. On your iPhone or iPad, localhost and 127.0.0.1 point to the mobile device and cannot reach your computer.

What next?

Try attaching an image with a vision-capable model, or learn how to organize your chats.

Connection did not work? Follow the checks for addresses, network access, and authentication.