Enter an instruction or select a preset, and watch the API respond with a completion that attempts to match the context or pattern you provided. You can control which model completes your request by changing the model.