fix: vllm gpu utilization

2025-11-27 00:50:42 +01:00
parent 5e9aa8f25d
commit eaa8e0ebab
2 changed files with 3 additions and 0 deletions
--- 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
--- 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