docs: update using "starter" distro than "ollama"

- update example
- set new env to use ollama
- use the same model as in llama-stack
- remove deprecated distro images

Signed-off-by: Wen Zhou <[email protected]>

Author: zdtsw

README.md

@@ -42,7 +42,7 @@ kubectl apply -f https://raw.githubusercontent.com/llamastack/llama-stack-k8s-op
 
 ### Deploying the Llama Stack Server
 
-1. Deploy the inference provider server (ollama, vllm etc). Example to deploy a new ollama server:
+1. Deploy the inference provider server (ollama). Example to deploy a new ollama server:
 ```
 bash hack/deploy-ollama.sh
 ```
@@ -56,14 +56,15 @@ spec:
   replicas: 1
   server:
     distribution:
-      name: ollama
+      name: starter
     containerSpec:
-      port: 8321
       env:
       - name: INFERENCE_MODEL
-        value: "llama3.2:1b"
+        value: "llama3.2:3b"
       - name: OLLAMA_URL
         value: "http://ollama-server-service.ollama-dist.svc.cluster.local:11434"
+      - name: ENABLE_OLLAMA
+        value: ollama
     storage:
       size: "20Gi"
       mountPath: "/home/lls/.lls"

config/manager/distribution-configmap.yaml

@@ -8,18 +8,7 @@ metadata:
 # The distribution images are immutable and should not be updated
 immutable: true
 data:
-  ollama: docker.io/llamastack/distribution-ollama:latest
-  hf-endpoint: docker.io/llamastack/distribution-hf-endpoint:latest
-  hf-serverless: docker.io/llamastack/distribution-hf-serverless:latest
-  bedrock: docker.io/llamastack/distribution-bedrock:latest
-  cerebras: docker.io/llamastack/distribution-cerebras:latest
-  nvidia: docker.io/llamastack/distribution-nvidia:latest
-  open-benchmark: docker.io/llamastack/distribution-open-benchmark:latest
-  passthrough: docker.io/llamastack/distribution-passthrough:latest
-  remote-vllm: docker.io/llamastack/distribution-remote-vllm:latest
-  sambanova: docker.io/llamastack/distribution-sambanova:latest
+  meta-reference-gpu: docker.io/llamastack/distribution-meta-reference-gpu:latest
   tgi: docker.io/llamastack/distribution-tgi:latest
-  together: docker.io/llamastack/distribution-together:latest
-  vllm-gpu: docker.io/llamastack/distribution-vllm-gpu:latest
-  watsonx: docker.io/llamastack/distribution-watsonx:latest
-  fireworks: docker.io/llamastack/distribution-fireworks:latest
+  postgres-demo: docker.io/llamastack/distribution-postgres-demo:latest
+  starter: docker.io/llamastack/distribution-starter:latest

config/samples/_v1alpha1_llamastackdistribution.yaml

@@ -8,12 +8,14 @@ spec:
     containerSpec:
       env:
         - name: INFERENCE_MODEL
-          value: 'llama3.2:1b'
+          value: 'llama3.2:3b'
         - name: OLLAMA_URL
           value: 'http://ollama-server-service.ollama-dist.svc.cluster.local:11434'
+        - name: ENABLE_OLLAMA
+          value: ollama
       name: llama-stack
     distribution:
-      name: ollama
+      name: starter
     # Uncomment the storage section to use persistent storage
     # storage: {}  # Will use default size of 10Gi and default mount path of /.llama
     # Or specify custom values:

config/samples/example-with-configmap.yaml

@@ -6,7 +6,7 @@ data:
   run.yaml: |
     # Llama Stack Configuration
     version: '2'
-    image_name: ollama
+    image_name: starter
     apis:
     - inference
     providers:
@@ -16,7 +16,7 @@ data:
         config:
           url: "http://ollama-server-service.ollama-dist.svc.cluster.local:11434"
     models:
-      - model_id: "llama3.2:1b"
+      - model_id: "ollama/llama3.2:3b"
         provider_id: ollama
         model_type: llm
     server:
@@ -30,14 +30,16 @@ spec:
   replicas: 1
   server:
     distribution:
-      name: ollama
+      name: starter
     containerSpec:
       port: 8321
       env:
       - name: INFERENCE_MODEL
-        value: "llama3.2:1b"
+        value: "llama3.2:3b"
       - name: OLLAMA_URL
         value: "http://ollama-server-service.ollama-dist.svc.cluster.local:11434"
+      - name: ENABLE_OLLAMA
+        value: ollama
     userConfig:
       configMapName: llama-stack-config
       # configMapNamespace: ""  # Optional - defaults to the same namespace as the CR

release/operator.yaml

@@ -2400,21 +2400,10 @@ subjects:
 ---
 apiVersion: v1
 data:
-  bedrock: docker.io/llamastack/distribution-bedrock:latest
-  cerebras: docker.io/llamastack/distribution-cerebras:latest
-  fireworks: docker.io/llamastack/distribution-fireworks:latest
-  hf-endpoint: docker.io/llamastack/distribution-hf-endpoint:latest
-  hf-serverless: docker.io/llamastack/distribution-hf-serverless:latest
-  nvidia: docker.io/llamastack/distribution-nvidia:latest
-  ollama: docker.io/llamastack/distribution-ollama:latest
-  open-benchmark: docker.io/llamastack/distribution-open-benchmark:latest
-  passthrough: docker.io/llamastack/distribution-passthrough:latest
-  remote-vllm: docker.io/llamastack/distribution-remote-vllm:latest
-  sambanova: docker.io/llamastack/distribution-sambanova:latest
+  meta-reference-gpu: docker.io/llamastack/distribution-meta-reference-gpu:latest
+  postgres-demo: docker.io/llamastack/distribution-postgres-demo:latest
+  starter: docker.io/llamastack/distribution-starter:latest
   tgi: docker.io/llamastack/distribution-tgi:latest
-  together: docker.io/llamastack/distribution-together:latest
-  vllm-gpu: docker.io/llamastack/distribution-vllm-gpu:latest
-  watsonx: docker.io/llamastack/distribution-watsonx:latest
 immutable: true
 kind: ConfigMap
 metadata: