version: 0.5.4-7
defaultConfig:
  namespace: llm
  model: Qwen/Qwen2.5-7B-Instruct
  maxModelLen: 8192
  gpuMemoryUtilization: 0.9
  cpuRequest: '4'
  cpuLimit: '8'
  memoryRequest: 16Gi
  memoryLimit: 24Gi
  gpuCount: 1
  domain: vllm.{{ .cloud.domain }}
defaultSecrets:
- key: apiKey