Update app.py

app.py

@@ -37,9 +37,9 @@ def load_quantized_model(model_id=None):
         )
         kwargs = {
             'model_path': model_path,
-            'n_ctx': 10000,
-            'max_tokens': 10000,
-            'n_batch': 512,
+            'n_ctx': 20000,
+            'max_tokens': 15000,
+            'n_batch': 1024,
             # 'n_gpu_layers':6,
         }
         return llamacpp.LlamaCpp(**kwargs)