Add Kubernetes support (#18)

deploy/kubernetes/kustomization.yaml

@@ -0,0 +1,20 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+- llama-gpt-api-deployment.yaml
+- llama-gpt-api-service.yaml
+- llama-gpt-ui-deployment.yaml
+- llama-gpt-ui-service.yaml
+
+# patches:
+# - 
+
+configMapGenerator:
+- name: llama-gpt
+  literals:
+  - DEFAULT_MODEL="/models/llama-2-7b-chat.bin"
+  - OPENAI_API_HOST="http://llama-gpt-api:8000"
+  - OPENAI_API_KEY="sk-XXXXXXXXXXXXXXXXXXXX"
+  - WAIT_HOSTS="llama-gpt-api:8000"
+  - WAIT_TIMEOUT="600"

deploy/kubernetes/llama-gpt-api-deployment.yaml

@@ -0,0 +1,29 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  labels:
+    service: llama-gpt-api
+  name: llama-gpt-api
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      service: llama-gpt-api
+  template:
+    metadata:
+      labels:
+        service: llama-gpt-api
+    spec:
+      containers:
+        - name: llama-gpt-api
+          image: ghcr.io/getumbrel/llama-gpt-api:1.0.1
+          env:
+            - name: MODEL
+              valueFrom: 
+                configMapKeyRef:
+                  name: llama-gpt
+                  key: DEFAULT_MODEL
+          resources:
+            requests:
+              memory: 5Gi
+      restartPolicy: Always

deploy/kubernetes/llama-gpt-api-service.yaml

@@ -0,0 +1,15 @@
+apiVersion: v1
+kind: Service
+metadata:
+  labels:
+    service: llama-gpt-api
+  name: llama-gpt-api
+spec:
+  ports:
+    - name: api
+      port: 8000
+      targetPort: 8000
+  selector:
+    service: llama-gpt-api
+status:
+  loadBalancer: {}

deploy/kubernetes/llama-gpt-ui-deployment.yaml

@@ -0,0 +1,26 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  labels:
+    service: llama-gpt-ui
+  name: llama-gpt-ui
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      service: llama-gpt-ui
+  template:
+    metadata:
+      labels:
+        service: llama-gpt-ui
+    spec:
+      containers:
+        - name: llama-gpt-ui
+          image: ghcr.io/getumbrel/llama-gpt-ui:latest
+          envFrom:
+          - configMapRef:
+              name: llama-gpt
+          ports:
+            - containerPort: 3000
+          resources: {}
+      restartPolicy: Always

deploy/kubernetes/llama-gpt-ui-service.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: Service
+metadata:
+  labels:
+    service: llama-gpt-ui
+  name: llama-gpt-ui
+spec:
+  ports:
+    - name: ui
+      port: 3000
+      targetPort: 3000
+  selector:
+    service: llama-gpt-ui
+  type: ClusterIP
+status:
+  loadBalancer: {}