diff --git a/gpt4all-backend/llmodel.h b/gpt4all-backend/llmodel.h
index 51bf4a23..5ecf9f3e 100644
--- a/gpt4all-backend/llmodel.h
+++ b/gpt4all-backend/llmodel.h
@@ -122,7 +122,7 @@ public:
         float   top_p = 0.9f;
         float   min_p = 0.0f;
         float   temp = 0.9f;
-        int32_t n_batch = 9;
+        int32_t n_batch = 128;
         float   repeat_penalty = 1.10f;
         int32_t repeat_last_n = 64;     // last n tokens to penalize
         float   contextErase = 0.75f;   // percent of context to erase if we exceed the context window