diff --git a/clusters/k3s-dgx/nim-service/qwen.yaml b/clusters/k3s-dgx/nim-service/qwen.yaml index aeb5088..6929e23 100644 --- a/clusters/k3s-dgx/nim-service/qwen.yaml +++ b/clusters/k3s-dgx/nim-service/qwen.yaml @@ -85,7 +85,7 @@ spec: type: PathPrefix value: / backendRefs: - - name: qwen3-32b-instruct + - name: qwen36-27b-fp8 port: 8000 timeouts: request: 600s