orign 0.2.3

A globally distributed container orchestrator
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: huggingface-cache-pvc
spec:
  storageClassName: gp2
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 100Gi
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: orign-vllm-{{model-name}}
spec:
  replicas: 1
  selector:
    matchLabels:
      app: orign-vllm-{{model-name}}
  template:
    metadata:
      labels:
        app: orign-vllm-{{model-name}}
    spec:
      hostIPC: true
      nodeSelector:
        role: gpu
        gpu-type: l40s-2xlarge
      containers:
        - name: orign-vllm-{{model-name}}
          image: us-docker.pkg.dev/agentsea-dev/orign/vllm:latest
          args: ["--model", "allenai/Molmo-7B-D-0924", "--dtype", "float32", "--trust-remote-code"]
          ports:
            - containerPort: 8000
          env:
            - name: HUGGING_FACE_HUB_TOKEN
              valueFrom:
                secretKeyRef:
                  name: huggingface-secret
                  key: HUGGING_FACE_HUB_TOKEN
          resources:
            limits:
              nvidia.com/gpu: 1
            requests:
              nvidia.com/gpu: 1
          volumeMounts:
            - name: huggingface-cache
              mountPath: /root/.cache/huggingface
      volumes:
        - name: huggingface-cache
          persistentVolumeClaim:
            claimName: huggingface-cache-pvc
---
apiVersion: v1
kind: Service
metadata:
  name: orign-vllm-{{model-name}}
spec:
  selector:
    app: orign-vllm
  ports:
    - protocol: TCP
      port: 8000
      targetPort: 8000
  type: LoadBalancer