AppGitHub
Getting Started

First Chat

Start your first conversation in Looma.

Create a Conversation

Open Looma and click New Conversation in the sidebar.

This creates a new chat session where Looma will manage the conversation context and execution.


Configure a Model

Before sending a message, you need to configure at least one model provider. Go to Settings → Models and add your provider.

Once the provider is connected, you will see a list of available models. Turn on the toggle next to the models you want to use.

Only enabled models will appear in the model selector in the chat input.

Looma supports both cloud and local models.

Common options include:

  • API-based providers (OpenAI, Anthropic, etc.)
  • Local model servers such as Ollama or LM Studio (the server must be running)

Once configured, select the model you want to use for the conversation.


Choose a Strategy

Every Looma conversation runs with a strategy. A strategy defines how the conversation works, including how context is built and how tools or memory may be used.

Looma includes several built-in strategies.

Examples include:

Standard Chat

A simple conversational strategy focused on natural dialogue.

Research Assistant

Optimized for gathering and summarizing information.

Memory Chat

Enables long-term knowledge storage and retrieval during conversations.

You can switch strategies at any time in the conversation settings.


Send Your First Message

Type a message in the input box and press Enter.

Looma will construct the conversation context using the selected strategy and send it to the model.

The response will appear in the chat window.

On this page