diff --git a/src/llama.cpp b/src/llama.cpp
index cb2fd5369..aeea54cff 100644
--- a/src/llama.cpp
+++ b/src/llama.cpp
@@ -17534,14 +17534,6 @@ struct llama_context * llama_new_context_with_model(
         params.flash_attn = false;
     }
 
-#ifdef GGML_USE_METAL
-    if (params.flash_attn && model->hparams.attn_soft_cap) {
-        LLAMA_LOG_WARN("%s: flash_attn is not compatible with attn_soft_cap - forcing off\n", __func__);
-        params.flash_attn = false;
-    }
-#endif // GGML_USE_METAL
-
-
     if (params.flash_attn && model->hparams.n_embd_head_k != model->hparams.n_embd_head_v) {
         LLAMA_LOG_WARN("%s: flash_attn requires n_embd_head_k == n_embd_head_v - forcing off\n", __func__);
         params.flash_attn = false;