diff --git a/llama.cpp b/llama.cpp
index 6f381f30f..db8e0e13d 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -1825,7 +1825,6 @@ static bool llama_eval_internal(
         //}
         ggml_metal_set_n_cb     (lctx.ctx_metal, n_threads);
         ggml_metal_graph_compute(lctx.ctx_metal, gf);
-        ggml_metal_get_tensor   (lctx.ctx_metal, cur);
     } else {
         // IMPORTANT:
         // Since we don't have efficient Matrix x Matrix Metal multiplication yet, we fallback to vanilla