Merge branch 'main' into patch-1

2026-01-01 20:50:00 +00:00 · 2025-01-17 23:49:05 -05:00 · 2025-01-17 23:49:05 -05:00 · f5edd07b29
commit f5edd07b29
parent 04934d4644 3a9468ce9b
91 changed files with 995 additions and 632 deletions
--- a/llama_stack/templates/bedrock/bedrock.py
+++ b/llama_stack/templates/bedrock/bedrock.py
@ -30,6 +30,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "bedrock"
@ -70,7 +71,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use AWS Bedrock for running LLM inference and safety",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=default_models,
--- a/llama_stack/templates/bedrock/build.yaml
+++ b/llama_stack/templates/bedrock/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/bedrock/run.yaml
+++ b/llama_stack/templates/bedrock/run.yaml
@ -81,6 +81,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/bedrock}/registry.db
--- a/llama_stack/templates/cerebras/cerebras.py
+++ b/llama_stack/templates/cerebras/cerebras.py
@ -92,7 +92,7 @@ def get_distribution_template() -> DistributionTemplate:
        name="cerebras",
        distro_type="self_hosted",
        description="Use Cerebras for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=default_models,
--- a/llama_stack/templates/experimental-post-training/build.yaml
+++ b/llama_stack/templates/experimental-post-training/build.yaml
@ -2,7 +2,7 @@ version: '2'
 name: experimental-post-training
 distribution_spec:
  description: Experimental template for post training
-  docker_image: null
+  container_image: null
  providers:
    inference:
    - inline::meta-reference
--- a/llama_stack/templates/experimental-post-training/run.yaml
+++ b/llama_stack/templates/experimental-post-training/run.yaml
@ -1,6 +1,6 @@
 version: '2'
 image_name: experimental-post-training
-docker_image: null
+container_image: null
 conda_env: experimental-post-training
 apis:
 - agents
--- a/llama_stack/templates/fireworks/build.yaml
+++ b/llama_stack/templates/fireworks/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/fireworks/fireworks.py
+++ b/llama_stack/templates/fireworks/fireworks.py
@ -39,6 +39,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }

@ -98,7 +99,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use Fireworks.AI for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=default_models,
--- a/llama_stack/templates/fireworks/run-with-safety.yaml
+++ b/llama_stack/templates/fireworks/run-with-safety.yaml
@ -92,6 +92,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db
--- a/llama_stack/templates/fireworks/run.yaml
+++ b/llama_stack/templates/fireworks/run.yaml
@ -86,6 +86,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db
--- a/llama_stack/templates/hf-endpoint/build.yaml
+++ b/llama_stack/templates/hf-endpoint/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/hf-endpoint/hf_endpoint.py
+++ b/llama_stack/templates/hf-endpoint/hf_endpoint.py
@ -34,6 +34,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "hf-endpoint"
@ -88,7 +89,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use (an external) Hugging Face Inference Endpoint for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=None,
        providers=providers,
        default_models=[inference_model, safety_model],
--- a/llama_stack/templates/hf-endpoint/run-with-safety.yaml
+++ b/llama_stack/templates/hf-endpoint/run-with-safety.yaml
@ -91,6 +91,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/registry.db
--- a/llama_stack/templates/hf-endpoint/run.yaml
+++ b/llama_stack/templates/hf-endpoint/run.yaml
@ -86,6 +86,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/registry.db
--- a/llama_stack/templates/hf-serverless/build.yaml
+++ b/llama_stack/templates/hf-serverless/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/hf-serverless/hf_serverless.py
+++ b/llama_stack/templates/hf-serverless/hf_serverless.py
@ -34,6 +34,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }

@ -89,7 +90,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use (an external) Hugging Face Inference Endpoint for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=None,
        providers=providers,
        default_models=[inference_model, safety_model],
--- a/llama_stack/templates/hf-serverless/run-with-safety.yaml
+++ b/llama_stack/templates/hf-serverless/run-with-safety.yaml
@ -91,6 +91,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/registry.db
--- a/llama_stack/templates/hf-serverless/run.yaml
+++ b/llama_stack/templates/hf-serverless/run.yaml
@ -86,6 +86,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/registry.db
--- a/llama_stack/templates/meta-reference-gpu/build.yaml
+++ b/llama_stack/templates/meta-reference-gpu/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/meta-reference-gpu/meta_reference.py
+++ b/llama_stack/templates/meta-reference-gpu/meta_reference.py
@ -38,6 +38,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "meta-reference-gpu"
--- a/llama_stack/templates/meta-reference-gpu/run-with-safety.yaml
+++ b/llama_stack/templates/meta-reference-gpu/run-with-safety.yaml
@ -93,6 +93,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/registry.db
--- a/llama_stack/templates/meta-reference-gpu/run.yaml
+++ b/llama_stack/templates/meta-reference-gpu/run.yaml
@ -87,6 +87,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/registry.db
--- a/llama_stack/templates/meta-reference-quantized-gpu/build.yaml
+++ b/llama_stack/templates/meta-reference-quantized-gpu/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py
+++ b/llama_stack/templates/meta-reference-quantized-gpu/meta_reference.py
@ -33,6 +33,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    default_tool_groups = [
--- a/llama_stack/templates/meta-reference-quantized-gpu/run.yaml
+++ b/llama_stack/templates/meta-reference-quantized-gpu/run.yaml
@ -89,6 +89,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-quantized-gpu}/registry.db
--- a/llama_stack/templates/nvidia/build.yaml
+++ b/llama_stack/templates/nvidia/build.yaml
@ -26,4 +26,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/nvidia/nvidia.py
+++ b/llama_stack/templates/nvidia/nvidia.py
@ -29,6 +29,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }

@ -68,7 +69,7 @@ def get_distribution_template() -> DistributionTemplate:
        name="nvidia",
        distro_type="remote_hosted",
        description="Use NVIDIA NIM for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=default_models,
--- a/llama_stack/templates/nvidia/run.yaml
+++ b/llama_stack/templates/nvidia/run.yaml
@ -83,6 +83,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/registry.db
--- a/llama_stack/templates/ollama/ollama.py
+++ b/llama_stack/templates/ollama/ollama.py
@ -90,7 +90,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use (an external) Ollama server for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=[inference_model, safety_model],
--- a/llama_stack/templates/remote-vllm/build.yaml
+++ b/llama_stack/templates/remote-vllm/build.yaml
@ -12,6 +12,15 @@ distribution_spec:
    - inline::llama-guard
    agents:
    - inline::meta-reference
+    eval:
+    - inline::meta-reference
+    datasetio:
+    - remote::huggingface
+    - inline::localfs
+    scoring:
+    - inline::basic
+    - inline::llm-as-judge
+    - inline::braintrust
    telemetry:
    - inline::meta-reference
    tool_runtime:
@ -19,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/remote-vllm/run-with-safety.yaml
+++ b/llama_stack/templates/remote-vllm/run-with-safety.yaml
@ -2,9 +2,12 @@ version: '2'
 image_name: remote-vllm
 apis:
 - agents
+- datasetio
+- eval
 - inference
 - memory
 - safety
+- scoring
 - telemetry
 - tool_runtime
 providers:
@ -44,6 +47,28 @@ providers:
        type: sqlite
        namespace: null
        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/agents_store.db
+  eval:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+  datasetio:
+  - provider_id: huggingface
+    provider_type: remote::huggingface
+    config: {}
+  - provider_id: localfs
+    provider_type: inline::localfs
+    config: {}
+  scoring:
+  - provider_id: basic
+    provider_type: inline::basic
+    config: {}
+  - provider_id: llm-as-judge
+    provider_type: inline::llm-as-judge
+    config: {}
+  - provider_id: braintrust
+    provider_type: inline::braintrust
+    config:
+      openai_api_key: ${env.OPENAI_API_KEY:}
  telemetry:
  - provider_id: meta-reference
    provider_type: inline::meta-reference
@ -68,6 +93,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/registry.db
--- a/llama_stack/templates/remote-vllm/run.yaml
+++ b/llama_stack/templates/remote-vllm/run.yaml
@ -2,9 +2,12 @@ version: '2'
 image_name: remote-vllm
 apis:
 - agents
+- datasetio
+- eval
 - inference
 - memory
 - safety
+- scoring
 - telemetry
 - tool_runtime
 providers:
@ -38,6 +41,28 @@ providers:
        type: sqlite
        namespace: null
        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/agents_store.db
+  eval:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+  datasetio:
+  - provider_id: huggingface
+    provider_type: remote::huggingface
+    config: {}
+  - provider_id: localfs
+    provider_type: inline::localfs
+    config: {}
+  scoring:
+  - provider_id: basic
+    provider_type: inline::basic
+    config: {}
+  - provider_id: llm-as-judge
+    provider_type: inline::llm-as-judge
+    config: {}
+  - provider_id: braintrust
+    provider_type: inline::braintrust
+    config:
+      openai_api_key: ${env.OPENAI_API_KEY:}
  telemetry:
  - provider_id: meta-reference
    provider_type: inline::meta-reference
@ -62,6 +87,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/registry.db
--- a/llama_stack/templates/remote-vllm/vllm.py
+++ b/llama_stack/templates/remote-vllm/vllm.py
@ -27,12 +27,16 @@ def get_distribution_template() -> DistributionTemplate:
        "memory": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
        "safety": ["inline::llama-guard"],
        "agents": ["inline::meta-reference"],
+        "eval": ["inline::meta-reference"],
+        "datasetio": ["remote::huggingface", "inline::localfs"],
+        "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::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "remote-vllm"
--- a/llama_stack/templates/template.py
+++ b/llama_stack/templates/template.py
@ -37,7 +37,7 @@ class RunConfigSettings(BaseModel):
        self,
        name: str,
        providers: Dict[str, List[str]],
-        docker_image: Optional[str] = None,
+        container_image: Optional[str] = None,
    ) -> StackRunConfig:
        provider_registry = get_provider_registry()

@ -83,7 +83,7 @@ class RunConfigSettings(BaseModel):

        return StackRunConfig(
            image_name=name,
-            docker_image=docker_image,
+            container_image=container_image,
            apis=apis,
            providers=provider_configs,
            metadata_store=SqliteKVStoreConfig.sample_run_config(
@ -112,7 +112,7 @@ class DistributionTemplate(BaseModel):

    # Optional configuration
    run_config_env_vars: Optional[Dict[str, Tuple[str, str]]] = None
-    docker_image: Optional[str] = None
+    container_image: Optional[str] = None

    default_models: Optional[List[ModelInput]] = None

@ -121,7 +121,7 @@ class DistributionTemplate(BaseModel):
            name=self.name,
            distribution_spec=DistributionSpec(
                description=self.description,
-                docker_image=self.docker_image,
+                container_image=self.container_image,
                providers=self.providers,
            ),
            image_type="conda",  # default to conda, can be overridden
@ -169,7 +169,7 @@ class DistributionTemplate(BaseModel):

        for yaml_pth, settings in self.run_configs.items():
            run_config = settings.run_config(
-                self.name, self.providers, self.docker_image
+                self.name, self.providers, self.container_image
            )
            with open(yaml_output_dir / yaml_pth, "w") as f:
                yaml.safe_dump(
--- a/llama_stack/templates/tgi/build.yaml
+++ b/llama_stack/templates/tgi/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/tgi/run-with-safety.yaml
+++ b/llama_stack/templates/tgi/run-with-safety.yaml
@ -86,6 +86,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/registry.db
--- a/llama_stack/templates/tgi/run.yaml
+++ b/llama_stack/templates/tgi/run.yaml
@ -85,6 +85,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/registry.db
--- a/llama_stack/templates/tgi/tgi.py
+++ b/llama_stack/templates/tgi/tgi.py
@ -36,6 +36,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "tgi"
@ -92,7 +93,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use (an external) TGI server for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=[inference_model, safety_model],
--- a/llama_stack/templates/together/build.yaml
+++ b/llama_stack/templates/together/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/together/run-with-safety.yaml
+++ b/llama_stack/templates/together/run-with-safety.yaml
@ -92,6 +92,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db
--- a/llama_stack/templates/together/run.yaml
+++ b/llama_stack/templates/together/run.yaml
@ -86,6 +86,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db
--- a/llama_stack/templates/together/together.py
+++ b/llama_stack/templates/together/together.py
@ -39,6 +39,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }
    name = "together"
@ -96,7 +97,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use Together.AI for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=Path(__file__).parent / "doc_template.md",
        providers=providers,
        default_models=default_models,
--- a/llama_stack/templates/vllm-gpu/build.yaml
+++ b/llama_stack/templates/vllm-gpu/build.yaml
@ -28,4 +28,5 @@ distribution_spec:
    - remote::tavily-search
    - inline::code-interpreter
    - inline::memory-runtime
+    - remote::model-context-protocol
 image_type: conda
--- a/llama_stack/templates/vllm-gpu/run.yaml
+++ b/llama_stack/templates/vllm-gpu/run.yaml
@ -89,6 +89,9 @@ providers:
  - provider_id: memory-runtime
    provider_type: inline::memory-runtime
    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
 metadata_store:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/vllm-gpu}/registry.db
--- a/llama_stack/templates/vllm-gpu/vllm.py
+++ b/llama_stack/templates/vllm-gpu/vllm.py
@ -33,6 +33,7 @@ def get_distribution_template() -> DistributionTemplate:
            "remote::tavily-search",
            "inline::code-interpreter",
            "inline::memory-runtime",
+            "remote::model-context-protocol",
        ],
    }

@ -84,7 +85,7 @@ def get_distribution_template() -> DistributionTemplate:
        name=name,
        distro_type="self_hosted",
        description="Use a built-in vLLM engine for running LLM inference",
-        docker_image=None,
+        container_image=None,
        template_path=None,
        providers=providers,
        default_models=[inference_model],