Update app.py

app.py

@@ -12,7 +12,7 @@ model_path = hf_hub_download(repo_id = repo_id, filename=model_file)
 # n_threads
 llm = Llama(model_path=model_path,n_gpu_layers=30,n_ctx=512,temperature=0.2,repeat_penalty=1.1,top_k_sampling=40,top_p_sampling=0.95,min_p_sampling=0.05)
 def generate_llm_response(prompt):
-    output = llm(prompt, max_tokens=512)
+    output = llm(prompt, max_tokens=256)
     return output["choices"][0]["text"]
 
 import streamlit as st