Update app.py

app.py

@@ -19,7 +19,7 @@ class ModelManager:
         self.model = None
         self.tokenizer = None
         self.setup_model()
-    
+
     def setup_model(self):
         try:
             self.tokenizer = AutoTokenizer.from_pretrained(
@@ -32,10 +32,23 @@ class ModelManager:
                 token=HF_TOKEN,
                 torch_dtype=torch.float16,
                 device_map="auto",
-                trust_remote_code=True
+                trust_remote_code=True,
+                low_cpu_mem_usage=True
             )
         except Exception as e:
-            raise Exception(f"Model loading failed: {e}")
+            print(f"Error loading model: {e}")
+            # Fallback to basic loading without device_map
+            try:
+                self.model = AutoModelForCausalLM.from_pretrained(
+                    MODEL_ID,
+                    token=HF_TOKEN,
+                    torch_dtype=torch.float16,
+                    trust_remote_code=True
+                )
+            except Exception as e:
+                raise Exception(f"Model loading failed completely: {e}")
+    
+    
 
 class ChatHistory:
     def __init__(self):