diff --git a/src/services/llm_local.rs b/src/services/llm_local.rs
index 4fd144f..264efaa 100644
--- a/src/services/llm_local.rs
+++ b/src/services/llm_local.rs
@@ -191,7 +191,7 @@ async fn start_llm_server(
 
     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 10000 --batch-size 10000 -n 4096 --mlock --no-mmap --prompt-cache-all --flash-attn  --no-kv-offload --tensor-split 100   --no-mmap &",
+        "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 &",
         llama_cpp_path, model_path, port
     ));