diff --git a/docs/source/distributions/remote_hosted_distro/nvidia.md b/docs/source/distributions/remote_hosted_distro/nvidia.md
index 3ebcbafff..f352f737e 100644
--- a/docs/source/distributions/remote_hosted_distro/nvidia.md
+++ b/docs/source/distributions/remote_hosted_distro/nvidia.md
@@ -9,7 +9,7 @@ The `llamastack/distribution-nvidia` distribution consists of the following prov
 | datasetio | `remote::huggingface`, `inline::localfs` |
 | eval | `inline::meta-reference` |
 | inference | `remote::nvidia` |
-| safety | `remote::fiddlecube` |
+| safety | `inline::llama-guard` |
 | scoring | `inline::basic`, `inline::llm-as-judge`, `inline::braintrust` |
 | telemetry | `inline::meta-reference` |
 | tool_runtime | `remote::brave-search`, `remote::tavily-search`, `inline::code-interpreter`, `inline::rag-runtime`, `remote::model-context-protocol` |
diff --git a/docs/source/distributions/self_hosted_distro/cerebras.md b/docs/source/distributions/self_hosted_distro/cerebras.md
index 05a9a6953..a0c9eb263 100644
--- a/docs/source/distributions/self_hosted_distro/cerebras.md
+++ b/docs/source/distributions/self_hosted_distro/cerebras.md
@@ -9,7 +9,7 @@ The `llamastack/distribution-cerebras` distribution consists of the following pr
 | datasetio | `remote::huggingface`, `inline::localfs` |
 | eval | `inline::meta-reference` |
 | inference | `remote::cerebras` |
-| safety | `remote::fiddlecube` |
+| safety | `inline::llama-guard` |
 | scoring | `inline::basic`, `inline::llm-as-judge`, `inline::braintrust` |
 | telemetry | `inline::meta-reference` |
 | tool_runtime | `remote::brave-search`, `remote::tavily-search`, `inline::code-interpreter`, `inline::rag-runtime` |
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 87a2c068f..9aeb7a88b 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
@@ -19,7 +19,7 @@ The `llamastack/distribution-meta-reference-quantized-gpu` distribution consists
 | datasetio | `remote::huggingface`, `inline::localfs` |
 | eval | `inline::meta-reference` |
 | inference | `inline::meta-reference-quantized` |
-| safety | `remote::fiddlecube` |
+| safety | `inline::llama-guard` |
 | scoring | `inline::basic`, `inline::llm-as-judge`, `inline::braintrust` |
 | telemetry | `inline::meta-reference` |
 | tool_runtime | `remote::brave-search`, `remote::tavily-search`, `inline::code-interpreter`, `inline::rag-runtime`, `remote::model-context-protocol` |
diff --git a/docs/source/distributions/self_hosted_distro/sambanova.md b/docs/source/distributions/self_hosted_distro/sambanova.md
index 02d89dfe3..e6ac616be 100644
--- a/docs/source/distributions/self_hosted_distro/sambanova.md
+++ b/docs/source/distributions/self_hosted_distro/sambanova.md
@@ -17,7 +17,7 @@ The `llamastack/distribution-sambanova` distribution consists of the following p
 |-----|-------------|
 | agents | `inline::meta-reference` |
 | inference | `remote::sambanova` |
-| safety | `remote::fiddlecube` |
+| safety | `inline::llama-guard` |
 | telemetry | `inline::meta-reference` |
 | tool_runtime | `remote::brave-search`, `remote::tavily-search`, `inline::code-interpreter`, `inline::rag-runtime` |
 | vector_io | `inline::faiss`, `remote::chromadb`, `remote::pgvector` |
diff --git a/llama_stack/templates/cerebras/build.yaml b/llama_stack/templates/cerebras/build.yaml
index 266b2d4b4..9d5ab1a52 100644
--- a/llama_stack/templates/cerebras/build.yaml
+++ b/llama_stack/templates/cerebras/build.yaml
@@ -5,7 +5,7 @@ distribution_spec:
     inference:
     - remote::cerebras
     safety:
-    - remote::fiddlecube
+    - inline::llama-guard
     vector_io:
     - inline::faiss
     - remote::chromadb
diff --git a/llama_stack/templates/cerebras/cerebras.py b/llama_stack/templates/cerebras/cerebras.py
index da0b79286..2dfae04f8 100644
--- a/llama_stack/templates/cerebras/cerebras.py
+++ b/llama_stack/templates/cerebras/cerebras.py
@@ -22,7 +22,7 @@ from llama_stack.templates.template import DistributionTemplate, RunConfigSettin
 def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["remote::cerebras"],
-        "safety": ["remote::fiddlecube"],
+        "safety": ["inline::llama-guard"],
         "vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
         "agents": ["inline::meta-reference"],
         "eval": ["inline::meta-reference"],
diff --git a/llama_stack/templates/cerebras/run.yaml b/llama_stack/templates/cerebras/run.yaml
index d0873357d..05d3f4525 100644
--- a/llama_stack/templates/cerebras/run.yaml
+++ b/llama_stack/templates/cerebras/run.yaml
@@ -21,8 +21,8 @@ providers:
     provider_type: inline::sentence-transformers
     config: {}
   safety:
-  - provider_id: fiddlecube
-    provider_type: remote::fiddlecube
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
     config: {}
   vector_io:
   - provider_id: faiss
diff --git a/llama_stack/templates/meta-reference-quantized-gpu/build.yaml b/llama_stack/templates/meta-reference-quantized-gpu/build.yaml
index 911f0f731..7bbcfe5f2 100644
--- a/llama_stack/templates/meta-reference-quantized-gpu/build.yaml
+++ b/llama_stack/templates/meta-reference-quantized-gpu/build.yaml
@@ -9,7 +9,7 @@ distribution_spec:
     - remote::chromadb
     - remote::pgvector
     safety:
-    - remote::fiddlecube
+    - inline::llama-guard
     agents:
     - inline::meta-reference
     telemetry:
diff --git a/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py b/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py
index fac46489a..8c2a6ec9f 100644
--- a/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py
+++ b/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py
@@ -22,7 +22,7 @@ def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["inline::meta-reference-quantized"],
         "vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
-        "safety": ["remote::fiddlecube"],
+        "safety": ["inline::llama-guard"],
         "agents": ["inline::meta-reference"],
         "telemetry": ["inline::meta-reference"],
         "eval": ["inline::meta-reference"],
diff --git a/llama_stack/templates/meta-reference-quantized-gpu/run.yaml b/llama_stack/templates/meta-reference-quantized-gpu/run.yaml
index e1152147a..d43cf3917 100644
--- a/llama_stack/templates/meta-reference-quantized-gpu/run.yaml
+++ b/llama_stack/templates/meta-reference-quantized-gpu/run.yaml
@@ -32,8 +32,8 @@ providers:
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-quantized-gpu}/faiss_store.db
   safety:
-  - provider_id: fiddlecube
-    provider_type: remote::fiddlecube
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
     config: {}
   agents:
   - provider_id: meta-reference
diff --git a/llama_stack/templates/nvidia/build.yaml b/llama_stack/templates/nvidia/build.yaml
index a05242f03..e9748721a 100644
--- a/llama_stack/templates/nvidia/build.yaml
+++ b/llama_stack/templates/nvidia/build.yaml
@@ -7,7 +7,7 @@ distribution_spec:
     vector_io:
     - inline::faiss
     safety:
-    - remote::fiddlecube
+    - inline::llama-guard
     agents:
     - inline::meta-reference
     telemetry:
diff --git a/llama_stack/templates/nvidia/nvidia.py b/llama_stack/templates/nvidia/nvidia.py
index eaf9ac500..d24c9ed48 100644
--- a/llama_stack/templates/nvidia/nvidia.py
+++ b/llama_stack/templates/nvidia/nvidia.py
@@ -18,7 +18,7 @@ def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["remote::nvidia"],
         "vector_io": ["inline::faiss"],
-        "safety": ["remote::fiddlecube"],
+        "safety": ["inline::llama-guard"],
         "agents": ["inline::meta-reference"],
         "telemetry": ["inline::meta-reference"],
         "eval": ["inline::meta-reference"],
diff --git a/llama_stack/templates/nvidia/run.yaml b/llama_stack/templates/nvidia/run.yaml
index 9bc572980..c8ae362f5 100644
--- a/llama_stack/templates/nvidia/run.yaml
+++ b/llama_stack/templates/nvidia/run.yaml
@@ -26,8 +26,8 @@ providers:
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/faiss_store.db
   safety:
-  - provider_id: fiddlecube
-    provider_type: remote::fiddlecube
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
     config: {}
   agents:
   - provider_id: meta-reference
diff --git a/llama_stack/templates/sambanova/build.yaml b/llama_stack/templates/sambanova/build.yaml
index 8a4e9c58b..ca5ffe618 100644
--- a/llama_stack/templates/sambanova/build.yaml
+++ b/llama_stack/templates/sambanova/build.yaml
@@ -9,7 +9,7 @@ distribution_spec:
     - remote::chromadb
     - remote::pgvector
     safety:
-    - remote::fiddlecube
+    - inline::llama-guard
     agents:
     - inline::meta-reference
     telemetry:
diff --git a/llama_stack/templates/sambanova/run.yaml b/llama_stack/templates/sambanova/run.yaml
index 56e122711..39b0f3c4e 100644
--- a/llama_stack/templates/sambanova/run.yaml
+++ b/llama_stack/templates/sambanova/run.yaml
@@ -29,8 +29,8 @@ providers:
     provider_type: remote::pgvector
     config: {}
   safety:
-  - provider_id: fiddlecube
-    provider_type: remote::fiddlecube
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
     config: {}
   agents:
   - provider_id: meta-reference
diff --git a/llama_stack/templates/sambanova/sambanova.py b/llama_stack/templates/sambanova/sambanova.py
index 9402624e4..70b54b010 100644
--- a/llama_stack/templates/sambanova/sambanova.py
+++ b/llama_stack/templates/sambanova/sambanova.py
@@ -24,7 +24,7 @@ def get_distribution_template() -> DistributionTemplate:
     providers = {
         "inference": ["remote::sambanova"],
         "vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
-        "safety": ["remote::fiddlecube"],
+        "safety": ["inline::llama-guard"],
         "agents": ["inline::meta-reference"],
         "telemetry": ["inline::meta-reference"],
         "tool_runtime": [