Update app.py

app.py

@@ -133,6 +133,7 @@ def generate_search_request():
         logits_all=True,
         #n_threads=12,
         verbose=True,
+        n_gpu_layers=20,
         n_gqa=8       #must be set for 70b models
     )
     
@@ -182,6 +183,7 @@ def generate_response():
         logits_all=True,
         #n_threads=12,
         verbose=True,
+        n_gpu_layers=20,
         n_gqa=8       #must be set for 70b models
     )