fix: vllm starter name

Signed-off-by: Sébastien Han <seb@redhat.com>

.github/workflows/providers-build.yml

@@ -10,6 +10,7 @@ on:
       - 'llama_stack/distribution/build.*'
       - 'llama_stack/distribution/*.sh'
       - '.github/workflows/providers-build.yml'
+      - 'llama_stack/templates/**'
   pull_request:
     paths:
       - 'llama_stack/cli/stack/build.py'
@@ -17,6 +18,7 @@ on:
       - 'llama_stack/distribution/build.*'
       - 'llama_stack/distribution/*.sh'
       - '.github/workflows/providers-build.yml'
+      - 'llama_stack/templates/**'
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}

llama_stack/templates/starter/build.yaml

@@ -11,7 +11,7 @@ distribution_spec:
     - remote::gemini
     - remote::groq
     - remote::sambanova
-    - remote::remote-vllm
+    - remote::vllm
     - inline::sentence-transformers
     vector_io:
     - inline::sqlite-vec

llama_stack/templates/starter/run.yaml

@@ -48,8 +48,8 @@ providers:
     config:
       url: https://api.sambanova.ai/v1
       api_key: ${env.SAMBANOVA_API_KEY:}
-  - provider_id: remote-vllm
-    provider_type: remote::remote-vllm
+  - provider_id: vllm
+    provider_type: remote::vllm
     config:
       url: ${env.VLLM_URL:http://localhost:8000/v1}
       max_tokens: ${env.VLLM_MAX_TOKENS:4096}

llama_stack/templates/starter/starter.py

@@ -108,7 +108,7 @@ def get_inference_providers() -> tuple[list[Provider], dict[str, list[ProviderMo
             SambaNovaImplConfig.sample_run_config(api_key="${env.SAMBANOVA_API_KEY:}"),
         ),
         (
-            "remote-vllm",
+            "vllm",
             [],
             VLLMInferenceAdapterConfig.sample_run_config(
                 url="${env.VLLM_URL:http://localhost:8000/v1}",