download_models.py

from transformers import (
    AutoProcessor,
    CLIPVisionModelWithProjection,
    CLIPTextModelWithProjection,
    AutoTokenizer,
    AutoImageProcessor,
    AutoModelForObjectDetection,
)

print('Downloading CLIP model...')
AutoProcessor.from_pretrained('openai/clip-vit-base-patch32', cache_dir='./models')
CLIPVisionModelWithProjection.from_pretrained('openai/clip-vit-base-patch32', cache_dir='./models')
CLIPTextModelWithProjection.from_pretrained('openai/clip-vit-base-patch32', cache_dir='./models')
AutoTokenizer.from_pretrained('openai/clip-vit-base-patch32', cache_dir='./models')

print('Downloading YOLO model...')
AutoImageProcessor.from_pretrained('hustvl/yolos-tiny', cache_dir='./models')
AutoModelForObjectDetection.from_pretrained('valentinafeve/yolos-fashionpedia', cache_dir='./models')

print('✅ All models downloaded!')