diff --git a/src/services/llm_local.rs b/src/services/llm_local.rs
index 264efaa..4e32c8b 100644
--- a/src/services/llm_local.rs
+++ b/src/services/llm_local.rs
@@ -189,9 +189,11 @@ async fn start_llm_server(
     std::env::set_var("OMP_PLACES", "cores");
     std::env::set_var("OMP_PROC_BIND", "close");
 
+    // "cd {} && numactl --interleave=all ./llama-server -m {} --host 0.0.0.0 --port {} --threads 20 --threads-batch 40 --temp 0.7 --parallel 1 --repeat-penalty 1.1 --ctx-size 8192 --batch-size 8192 -n 4096 --mlock --no-mmap --flash-attn  --no-kv-offload  --no-mmap &",
+
     let mut cmd = tokio::process::Command::new("sh");
     cmd.arg("-c").arg(format!(
-        "cd {} && numactl --interleave=all ./llama-server -m {} --host 0.0.0.0 --port {} --threads 20 --threads-batch 40 --temp 0.7 --parallel 1 --repeat-penalty 1.1 --ctx-size 8192 --batch-size 8192 -n 4096 --mlock --no-mmap --flash-attn  --no-kv-offload  --no-mmap &",
+        "cd {} && ./llama-server -m {} --host 0.0.0.0 --port {} --n-gpu-layers 99 --mlock --no-mmap --threads 20 --threads-batch 40 &",
         llama_cpp_path, model_path, port
     ));