vllm deploy

2026-05-05 01:03:33 +00:00 · 2026-05-05 01:03:33 +00:00 · e0971380ad
commit e0971380ad
parent 4f82d918bd
1 changed files with 2 additions and 2 deletions
--- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
+++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
@ -59,8 +59,8 @@ spec:
            requests:
              nvidia.com/gpu: 1
-              memory: "250Gi"
+              memory: "150Gi"
-              cpu: "40"
+              cpu: "24"
          startupProbe:
            httpGet: