Create model_loader.py

utils/model_loader.py

@@ -0,0 +1,18 @@
+from transformers import pipeline
+import torch
+
+def load_llava_model():
+    return pipeline(
+        "image-to-text",
+        model="llava-hf/llava-1.5-7b-hf",
+        torch_dtype=torch.float16,
+        device_map="auto",
+        max_new_tokens=200
+    )
+
+def load_clip_model():
+    return pipeline(
+        "feature-extraction",
+        model="openai/clip-vit-base-patch32",
+        device_map="auto"
+    )