Three LoRA models in "Next Token Is Enough: Realistic Image Quality and Aesthetic Scoring with Multimodal Large Language Model"
-