diff --git a/src-tauri/src/core/utils/extensions/inference_llamacpp_extension/server.rs b/src-tauri/src/core/utils/extensions/inference_llamacpp_extension/server.rs
index 449e0e543..8e2d182cd 100644
--- a/src-tauri/src/core/utils/extensions/inference_llamacpp_extension/server.rs
+++ b/src-tauri/src/core/utils/extensions/inference_llamacpp_extension/server.rs
@@ -144,8 +144,9 @@ pub async fn load_llama_model(
     #[cfg(all(windows, target_arch = "x86_64"))]
     {
         use std::os::windows::process::CommandExt;
+        const CREATE_NO_WINDOW: u32 = 0x0800_0000;
         const CREATE_NEW_PROCESS_GROUP: u32 = 0x0000_0200;
-        command.creation_flags(CREATE_NEW_PROCESS_GROUP);
+        command.creation_flags(CREATE_NO_WINDOW | CREATE_NEW_PROCESS_GROUP);
     }
 
     // Spawn the child process