Update app.py

app.py

@@ -62,8 +62,7 @@ def respond(
     prompt += f"<user>{message}</user>"
 
     # Tokenize the input prompt
-    inputs = tokenizer(prompt, return_tensors="pt", truncation=True).to(device)
-
+    inputs = tokenizer(prompt, return_tensors="pt", truncation=True)
     # Generate the output
     outputs = peft_model.generate(
         **inputs,