Update app.py

app.py

@@ -97,14 +97,13 @@ def mode_load(path):
 
 
 @spaces.GPU()
-
-model = AutoModelForCausalLM.from_pretrained(
-    MODEL_ID,
-    torch_dtype=torch.bfloat16,
-    low_cpu_mem_usage=True,
-    trust_remote_code=True
-)
-model.eval()
+    model = AutoModelForCausalLM.from_pretrained(
+        MODEL_ID,
+        torch_dtype=torch.bfloat16,
+        low_cpu_mem_usage=True,
+        trust_remote_code=True
+    )
+    model.eval()
 def stream_chat(message, history: list, temperature: float, max_length: int, top_p: float, top_k: int, penalty: float):
     print(f'message is - {message}')
     print(f'history is - {history}')