version: 0.9.2-1
defaultConfig:
  namespace: llm
  model: Qwen/Qwen3-8B
  maxModelLen: 32768
  gpuMemoryUtilization: 0.9
  cpuRequest: '4'
  cpuLimit: '8'
  memoryRequest: 16Gi
  memoryLimit: 24Gi
  gpuCount: 1
  domain: vllm.{{ .cloud.domain }}
defaultSecrets:
- key: apiKey