diff --git a/python_server/__pycache__/server.cpython-313.pyc b/python_server/__pycache__/server.cpython-313.pyc
index a70f5f9..7e87f12 100644
Binary files a/python_server/__pycache__/server.cpython-313.pyc and b/python_server/__pycache__/server.cpython-313.pyc differ
diff --git a/python_server/server.py b/python_server/server.py
index b542839..b2091ee 100644
--- a/python_server/server.py
+++ b/python_server/server.py
@@ -41,16 +41,15 @@ logging.basicConfig(level=logging.INFO)
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 logging.info(f"Using device: {device}")
 
-# Adjust this path for local model folder
-#model_path = "./model/clip-vit-base-patch32"
-model_path = "models/clip-vit-large-patch14"  
+# Use model directly from HuggingFace (auto-downloads and caches)
+model_path = "openai/clip-vit-base-patch32"  # Using base model for faster loading
 
-
-logging.info("Loading CLIP model...")
+logging.info(f"Loading CLIP model from HuggingFace: {model_path}...")
+logging.info("(Model will be cached after first download)")
 model = CLIPModel.from_pretrained(model_path).to(device)
 processor = CLIPProcessor.from_pretrained(model_path)
 model.eval()
-logging.info("Model loaded.")
+logging.info("Model loaded successfully.")
 
 # Qdrant connection
 '''qdrant = QdrantClient(