By default, Banani uses Auto mode, which selects the best model for each prompt.
For example, when generating a new complex screen, it will use a higher reasoning model like Gemini 3.1 Pro. When making edits, it switches to faster models like GPT-5.4 with lower reasoning.
The good thing is you can also always manually choose which model to use.
If you're creating a new project
Click "Auto" with the Settings icon at the bottom left of the prompt input
Click Model
Select the desired model from the list
Inside an existing project
Click "Auto" with the lightning and spark icon
Select a model from the list
Sometimes we experience extended request times, so if your generation fails, try switching to a different model.