This model was trained in the 1.5 model, its purpose is to help in the archviz area
The LoRa model works with the trigger word, but sometimes it is required