| # Load model directly | |
| from transformers import AutoProcessor, AutoModelForImageTextToText | |
| processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM-500M-Instruct") | |
| model = AutoModelForImageTextToText.from_pretrained("HuggingFaceTB/SmolVLM-500M-Instruct") | |