inferenceExtension:
  pluginsConfigFile: default-plugins.yaml
  routing:
    deploymentMode: pd
    profile: prediction
    pd:
      pdLabelName: openfuyao.com/pdRole
      pdGroupLabelName: openfuyao.com/pdGroupID
      prefillValue: prefill
      decodeValue: decode
      leaderValue: leader
    tokenizer:
      model: Qwen/Qwen3-8B
      socketPath: /var/run/tokenizer/tokenizer.sock
    requestTracking:
      storeName: hermes-inflight
      persistence:
        enabled: false
        flushThreshold: 100
        outputPath: /tmp/hermes-inflight/completed.jsonl
    cacheIndexer:
      address: http://cache-indexer-service:8080
    npuExporter:
      exporterNamespace: npu-exporter
      exporterPodLabelSelector: app=npu-exporter
      exporterPort: 8082
      path: /metrics
      cacheTTL: 500ms
      scrapeTimeout: 1s
      staleAfter: 5s
  prediction:
    enabled: true
    # Set targetModel/modelVersion/modelVolume to load a real bundle.
    # examples ships only an aggregate bundle — train your own for pd.
    socketPath: /var/run/hermes/prediction.sock
    timeout: 1s
    maxBatchSize: 128