Update app.py

app.py

@@ -24,7 +24,6 @@ def generate_solution(input_text):
     print(input_text)
     
     input_ids = tokenizer(input_text, return_tensors='pt', padding="max_length", truncation=True, max_length=128).input_ids
-    input_ids = input_ids.to(model.device)
 
     with torch.no_grad():
         outputs = quantized_model.generate(input_ids, max_length=128, num_beams=4, early_stopping=True)