Update app.py

app.py

@@ -43,8 +43,11 @@ def chat(history, system_message, max_tokens, temperature, top_p, top_k, repetit
     # Apply the chat template
     gen_input = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
     
+    # Extract input_ids
+    input_ids = gen_input['input_ids']
+    
     # Generate the output
-    output = model.generate(input_ids=gen_input.input_ids, temperature=temperature, do_sample=True, top_p=top_p, top_k=top_k, max_length=max_tokens)
+    output = model.generate(input_ids=input_ids, temperature=temperature, do_sample=True, top_p=top_p, top_k=top_k, max_length=max_tokens)
     
     # Decode the output
     decoded_output = tokenizer.decode(output[0], skip_special_tokens=True)
@@ -54,6 +57,7 @@ def chat(history, system_message, max_tokens, temperature, top_p, top_k, repetit
     
     return history, history, ""
 
+
 start_message = ""
 
 CSS ="""