29 lines
501 B
YAML
29 lines
501 B
YAML
---
|
|
vllm_chart_version: "0.1.0"
|
|
vllm_namespace: vllm-system
|
|
|
|
vllm_deployment_mode: "distributed"
|
|
vllm_num_replicas: 2
|
|
|
|
vllm_image:
|
|
repository: "vllm/vllm-openai"
|
|
tag: "latest"
|
|
pull_policy: "IfNotPresent"
|
|
|
|
vllm_resources:
|
|
limits:
|
|
nvidia.com/gpu: "1"
|
|
requests:
|
|
nvidia.com/gpu: "1"
|
|
|
|
vllm_env:
|
|
- name: "NGC_API_KEY"
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: "vllm-ngc-secret"
|
|
key: "api-key"
|
|
|
|
vllm_extra_args: []
|
|
|
|
vllm_kubeconfig: "/home/{{ ansible_user }}/.kube/config"
|