ai-llm: GPU qwen2.5-coder on titan-24; add chat.ai host

services/ai-llm/deployment.yaml

@@ -35,7 +35,7 @@ spec:
             - name: OLLAMA_MODELS
               value: /root/.ollama
             - name: OLLAMA_MODEL
-              value: phi3:mini
+              value: qwen2.5-coder:7b-instruct-q4_0
           command:
             - /bin/sh
             - -c
@@ -52,6 +52,9 @@ spec:
             requests:
               cpu: 250m
               memory: 1Gi
+              nvidia.com/gpu: 1
+            limits:
+              nvidia.com/gpu: 1
       containers:
         - name: ollama
           image: ollama/ollama:latest
@@ -77,6 +80,8 @@ spec:
             requests:
               cpu: "2"
               memory: 8Gi
+              nvidia.com/gpu: 1
             limits:
               cpu: "4"
               memory: 12Gi
+              nvidia.com/gpu: 1

services/bstein-dev-home/ingress.yaml

@@ -9,9 +9,9 @@ metadata:
     traefik.ingress.kubernetes.io/router.entrypoints: websecure
     traefik.ingress.kubernetes.io/router.tls: "true"
     cert-manager.io/cluster-issuer: letsencrypt
-spec:
+  spec:
   tls:
-    - hosts: [ "bstein.dev" ]
+    - hosts: [ "bstein.dev", "chat.ai.bstein.dev" ]
       secretName: bstein-dev-home-tls
   rules:
     - host: bstein.dev
@@ -29,3 +29,18 @@ spec:
               service:
                 name: bstein-dev-home-frontend
                 port: { number: 80 }
+    - host: chat.ai.bstein.dev
+      http:
+        paths:
+          - path: /api
+            pathType: Prefix
+            backend:
+              service:
+                name: bstein-dev-home-backend
+                port: { number: 80 }
+          - path: /
+            pathType: Prefix
+            backend:
+              service:
+                name: bstein-dev-home-frontend
+                port: { number: 80 }