Update base model id

app.py

@@ -13,14 +13,17 @@ import os
 # Load your custom VLM model from Hugging Face
 # Replace with your actual model identifier
 MODEL_ID = "enpeizhao/qwen2_5-3b-instruct-trl-sft-vlm-odd-12-nf4-merged"  
+BASE_MODEL_ID = "Qwen/Qwen2.5-3B-Instruct"  
+
+
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 def load_model():
     """Load the model and processor from Hugging Face"""
     model = AutoModel.from_pretrained(MODEL_ID, trust_remote_code=True).to(device)
-    processor = AutoProcessor.from_pretrained(MODEL_ID, trust_remote_code=True)
-    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
+    processor = AutoProcessor.from_pretrained(BASE_MODEL_ID, trust_remote_code=True)
+    tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL_ID, trust_remote_code=True)
     return model, processor, tokenizer
 
 # Load model at startup