diff --git a/docs/source/distributions/self_hosted_distro/meta-reference-gpu.md b/docs/source/distributions/self_hosted_distro/meta-reference-gpu.md
index f6df8b1f3..3fcb16e12 100644
--- a/docs/source/distributions/self_hosted_distro/meta-reference-gpu.md
+++ b/docs/source/distributions/self_hosted_distro/meta-reference-gpu.md
@@ -96,5 +96,5 @@ If you are using Llama Stack Safety / Shield APIs, use:
 llama stack run distributions/meta-reference-gpu/run-with-safety.yaml \
   --port 5001 \
   --env INFERENCE_MODEL=Llama3.2-3B-Instruct \
-  --env SAFETY_MODEL=meta-Llama-Guard-3-1B
+  --env SAFETY_MODEL=Llama-Guard-3-1B
 ```
diff --git a/docs/source/distributions/self_hosted_distro/meta-reference-quantized-gpu.md b/docs/source/distributions/self_hosted_distro/meta-reference-quantized-gpu.md
index 73db33026..23302a3ab 100644
--- a/docs/source/distributions/self_hosted_distro/meta-reference-quantized-gpu.md
+++ b/docs/source/distributions/self_hosted_distro/meta-reference-quantized-gpu.md
@@ -76,7 +76,7 @@ docker run \
   llamastack/distribution-meta-reference-quantized-gpu \
   --port $LLAMA_STACK_PORT \
   --env INFERENCE_MODEL=Llama3.2-3B-Instruct \
-  --env SAFETY_MODEL=meta-Llama-Guard-3-1B
+  --env SAFETY_MODEL=Llama-Guard-3-1B
 ```
 
 ### Via Conda
diff --git a/llama_stack/providers/inline/inference/meta_reference/__init__.py b/llama_stack/providers/inline/inference/meta_reference/__init__.py
index 2ae3e7a16..9c923490d 100644
--- a/llama_stack/providers/inline/inference/meta_reference/__init__.py
+++ b/llama_stack/providers/inline/inference/meta_reference/__init__.py
@@ -17,5 +17,4 @@ async def get_provider_impl(
 
     impl = MetaReferenceInferenceImpl(config)
     await impl.initialize()
-
     return impl