diff --git a/.github/workflows/integration-tests.yml b/.github/workflows/integration-tests.yml
index e6d0475d3..098f511cb 100644
--- a/.github/workflows/integration-tests.yml
+++ b/.github/workflows/integration-tests.yml
@@ -101,6 +101,11 @@ jobs:
 
       - name: Run Integration Tests
         env:
+          OLLAMA_INFERENCE_MODEL: "llama3.2:3b-instruct-fp16" # for server tests
+          ENABLE_OLLAMA: "ollama" # for server tests
+          OLLAMA_URL: "http://0.0.0.0:11434"
+          SAFETY_MODEL: "llama-guard3:1b"
+          ENABLE_HUGGINGFACE: "huggingface"
           LLAMA_STACK_CLIENT_TIMEOUT: "300" # Increased timeout for eval operations
         # Use 'shell' to get pipefail behavior
         # https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#exit-codes-and-error-action-preference
diff --git a/llama_stack/templates/starter/run.yaml b/llama_stack/templates/starter/run.yaml
index c38933f98..6bceee6bc 100644
--- a/llama_stack/templates/starter/run.yaml
+++ b/llama_stack/templates/starter/run.yaml
@@ -179,7 +179,7 @@ providers:
       sqlite_db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/starter}/trace_store.db
       otel_exporter_otlp_endpoint: ${env.OTEL_EXPORTER_OTLP_ENDPOINT:=}
   post_training:
-  - provider_id: huggingface
+  - provider_id: ${env.ENABLE_HUGGINGFACE:=__disabled__}
     provider_type: inline::huggingface
     config:
       checkpoint_format: huggingface
diff --git a/llama_stack/templates/starter/starter.py b/llama_stack/templates/starter/starter.py
index cee1094db..e31c02e9b 100644
--- a/llama_stack/templates/starter/starter.py
+++ b/llama_stack/templates/starter/starter.py
@@ -281,7 +281,7 @@ def get_distribution_template() -> DistributionTemplate:
         config=SentenceTransformersInferenceConfig.sample_run_config(),
     )
     post_training_provider = Provider(
-        provider_id="huggingface",
+        provider_id="${env.ENABLE_HUGGINGFACE:=__disabled__}",
         provider_type="inline::huggingface",
         config=HuggingFacePostTrainingConfig.sample_run_config(f"~/.llama/distributions/{name}"),
     )