diff --git a/llama.cpp b/llama.cpp
index 8c89e9912e9faf..bd065e961adecc 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -136,7 +136,7 @@ struct llama_kv_cache {
 
     struct ggml_context * ctx = NULL;
 
-    llama_buffer buf;
+    llama_ctx_buffer buf;
 
     int n; // number of tokens currently in the cache