docker run -d --name llama-server \
  -p 8080:8080 \
  -v $(pwd)/models:/models \
  docker.io/samueltallet/alpine-llama-cpp-server:latest \
  --model /models/llama-2-7b.Q4_K_M.gguf --host 0.0.0.0 --port 8080

docker run -d --name llama-server \
  --restart unless-stopped \
  -p 8080:8080 \
  -v /data/models:/models:ro \
  docker.io/samueltallet/alpine-llama-cpp-server:0.3.0 \
  --model /models/llama-2-7b.Q4_K_M.gguf \
  --host 0.0.0.0 --port 8080 --ctx-size 4096 --parallel 2

apiVersion: apps/v1
kind: Deployment
metadata:
  name: llama-cpp-server
spec:
  replicas: 2
  selector:
    matchLabels:
      app: llama-cpp-server
  template:
    metadata:
      labels:
        app: llama-cpp-server
    spec:
      containers:
        - name: server
          image: docker.io/samueltallet/alpine-llama-cpp-server:0.3.0
          args:
            - --model
            - /models/llama-2-7b.Q4_K_M.gguf
            - --host
            - 0.0.0.0
            - --port
            - "8080"
          ports:
            - containerPort: 8080
          resources:
            requests:
              memory: 8Gi
          volumeMounts:
            - name: models
              mountPath: /models
              readOnly: true
      volumes:
        - name: models
          persistentVolumeClaim:
            claimName: llama-models
---
apiVersion: v1
kind: Service
metadata:
  name: llama-cpp-server
spec:
  selector:
    app: llama-cpp-server
  ports:
    - port: 8080
      targetPort: 8080