Use the Model Selector

In Optic, we use a feature called the Model Selector to choose which model to use for the current conversation.

Level 1: Models

At the top of the Optic app, there is a button labeled “Select a model.” Clicking it opens a dialog box—this is the model selector.

You'll see an empty list of models because you haven't added any yet. Click “Add Model” to proceed to Level 2.

Level 2: LLM Services

To add a model, you need to select an LLM service. We've pre-installed some services; selecting one will take you to Level 3. Alternatively, you can add a custom service. When you click “Add LLM Service,” the new service will appear directly in the list.

Level 3: All Models from the LLM Service

Let's say you selected OpenAI. You'll now see options to configure the LLM service and an empty list of models. Click the configuration option, and you'll be prompted to enter an API key. This dialog also provides a link to quickly open the OpenAI platform.

As we mentioned earlier: open the OpenAI platform, create an API key, and enter it here—it's that simple.

After entering the API Key, you will be redirected to the Level 3 interface, where Optic will automatically begin fetching all models. Simply click on the model you want to add, such as gpt-5.5 or gpt-5.4-mini. The model will then be successfully added.

For faster navigation, you can use keyboard shortcuts to filter models: press the up and down arrow keys to navigate, and press Enter to add a model.

Once you've added your desired models, press the Back button or the Esc key to return to Level 1, then select the model you want to use for the current conversation. You're all set.