diff --git a/services/vllm/config_bge.yaml b/services/vllm/config_bge.yaml
index cae1465..3b45c3d 100644
--- a/services/vllm/config_bge.yaml
+++ b/services/vllm/config_bge.yaml
@@ -2,3 +2,4 @@ model: BAAI/bge-large-en-v1.5
 host: "0.0.0.0"
 port: 8002
 uvicorn-log-level: "info"
+gpu-memory-utilization: 0.25
diff --git a/services/vllm/config_llama.yaml b/services/vllm/config_llama.yaml
index ddc68a1..7c824ef 100644
--- a/services/vllm/config_llama.yaml
+++ b/services/vllm/config_llama.yaml
@@ -2,3 +2,5 @@ model: meta-llama/Llama-3.1-8B-Instruct
 host: "0.0.0.0"
 port: 8001
 uvicorn-log-level: "info"
+gpu-memory-utilization: 0.75
+max-model-len: 20480