Update app.py

app.py

@@ -33,7 +33,7 @@ tokenizer = AutoTokenizer.from_pretrained(model_repo, use_fast=False)
 
 def generate_response(prompt, max_tokens=128, temperature=0.7, top_p=0.9):
     try:
-        chat = [{"role": "user", "content": user_input}]
+        chat = [{"role": "user", "content": prompt}]
         prompt = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
 
         inputs = tokenizer(prompt, return_tensors="pt").to(model.device)