Merge 76da90c1fc into sapling-pr-archive-ehhuang

llama_stack/core/datatypes.py

@@ -422,6 +422,18 @@ def process_cors_config(cors_config: bool | CORSConfig | None) -> CORSConfig | N
     raise ValueError(f"Expected bool or CORSConfig, got {type(cors_config).__name__}")
 
 
+class RegisteredResources(BaseModel):
+    """Registry of resources available in the distribution."""
+
+    models: list[ModelInput] = Field(default_factory=list)
+    shields: list[ShieldInput] = Field(default_factory=list)
+    vector_dbs: list[VectorDBInput] = Field(default_factory=list)
+    datasets: list[DatasetInput] = Field(default_factory=list)
+    scoring_fns: list[ScoringFnInput] = Field(default_factory=list)
+    benchmarks: list[BenchmarkInput] = Field(default_factory=list)
+    tool_groups: list[ToolGroupInput] = Field(default_factory=list)
+
+
 class ServerConfig(BaseModel):
     port: int = Field(
         default=8321,
@@ -491,14 +503,10 @@ can be instantiated multiple times (with different configs) if necessary.
         description="Catalog of named storage backends and references available to the stack",
     )
 
-    # registry of "resources" in the distribution
-    models: list[ModelInput] = Field(default_factory=list)
-    shields: list[ShieldInput] = Field(default_factory=list)
-    vector_dbs: list[VectorDBInput] = Field(default_factory=list)
-    datasets: list[DatasetInput] = Field(default_factory=list)
-    scoring_fns: list[ScoringFnInput] = Field(default_factory=list)
-    benchmarks: list[BenchmarkInput] = Field(default_factory=list)
-    tool_groups: list[ToolGroupInput] = Field(default_factory=list)
+    registered_resources: RegisteredResources = Field(
+        default_factory=RegisteredResources,
+        description="Registry of resources available in the distribution",
+    )
 
     logging: LoggingConfig | None = Field(default=None, description="Configuration for Llama Stack Logging")
 

llama_stack/core/stack.py

@@ -110,7 +110,7 @@ TEST_RECORDING_CONTEXT = None
 
 async def register_resources(run_config: StackRunConfig, impls: dict[Api, Any]):
     for rsrc, api, register_method, list_method in RESOURCES:
-        objects = getattr(run_config, rsrc)
+        objects = getattr(run_config.registered_resources, rsrc)
         if api not in impls:
             continue
 

llama_stack/distributions/ci-tests/run.yaml

@@ -247,23 +247,24 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models: []
-shields:
-- shield_id: llama-guard
-  provider_id: ${env.SAFETY_MODEL:+llama-guard}
-  provider_shield_id: ${env.SAFETY_MODEL:=}
-- shield_id: code-scanner
-  provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
-  provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models: []
+  shields:
+  - shield_id: llama-guard
+    provider_id: ${env.SAFETY_MODEL:+llama-guard}
+    provider_shield_id: ${env.SAFETY_MODEL:=}
+  - shield_id: code-scanner
+    provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
+    provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/dell/run-with-safety.yaml

@@ -109,31 +109,32 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: tgi0
-  model_type: llm
-- metadata: {}
-  model_id: ${env.SAFETY_MODEL}
-  provider_id: tgi1
-  model_type: llm
-- metadata:
-    embedding_dimension: 768
-  model_id: nomic-embed-text-v1.5
-  provider_id: sentence-transformers
-  model_type: embedding
-shields:
-- shield_id: ${env.SAFETY_MODEL}
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: brave-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: tgi0
+    model_type: llm
+  - metadata: {}
+    model_id: ${env.SAFETY_MODEL}
+    provider_id: tgi1
+    model_type: llm
+  - metadata:
+      embedding_dimension: 768
+    model_id: nomic-embed-text-v1.5
+    provider_id: sentence-transformers
+    model_type: embedding
+  shields:
+  - shield_id: ${env.SAFETY_MODEL}
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: brave-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/dell/run.yaml

@@ -105,26 +105,27 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: tgi0
-  model_type: llm
-- metadata:
-    embedding_dimension: 768
-  model_id: nomic-embed-text-v1.5
-  provider_id: sentence-transformers
-  model_type: embedding
-shields: []
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: brave-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: tgi0
+    model_type: llm
+  - metadata:
+      embedding_dimension: 768
+    model_id: nomic-embed-text-v1.5
+    provider_id: sentence-transformers
+    model_type: embedding
+  shields: []
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: brave-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/meta-reference-gpu/run-with-safety.yaml

@@ -122,31 +122,32 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: meta-reference-inference
-  model_type: llm
-- metadata: {}
-  model_id: ${env.SAFETY_MODEL}
-  provider_id: meta-reference-safety
-  model_type: llm
-- metadata:
-    embedding_dimension: 768
-  model_id: nomic-embed-text-v1.5
-  provider_id: sentence-transformers
-  model_type: embedding
-shields:
-- shield_id: ${env.SAFETY_MODEL}
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: meta-reference-inference
+    model_type: llm
+  - metadata: {}
+    model_id: ${env.SAFETY_MODEL}
+    provider_id: meta-reference-safety
+    model_type: llm
+  - metadata:
+      embedding_dimension: 768
+    model_id: nomic-embed-text-v1.5
+    provider_id: sentence-transformers
+    model_type: embedding
+  shields:
+  - shield_id: ${env.SAFETY_MODEL}
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/meta-reference-gpu/run.yaml

@@ -112,26 +112,27 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: meta-reference-inference
-  model_type: llm
-- metadata:
-    embedding_dimension: 768
-  model_id: nomic-embed-text-v1.5
-  provider_id: sentence-transformers
-  model_type: embedding
-shields: []
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: meta-reference-inference
+    model_type: llm
+  - metadata:
+      embedding_dimension: 768
+    model_id: nomic-embed-text-v1.5
+    provider_id: sentence-transformers
+    model_type: embedding
+  shields: []
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/nvidia/run-with-safety.yaml

@@ -111,25 +111,26 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: nvidia
-  model_type: llm
-- metadata: {}
-  model_id: ${env.SAFETY_MODEL}
-  provider_id: nvidia
-  model_type: llm
-shields:
-- shield_id: ${env.SAFETY_MODEL}
-  provider_id: nvidia
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: nvidia
+    model_type: llm
+  - metadata: {}
+    model_id: ${env.SAFETY_MODEL}
+    provider_id: nvidia
+    model_type: llm
+  shields:
+  - shield_id: ${env.SAFETY_MODEL}
+    provider_id: nvidia
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/nvidia/run.yaml

@@ -100,15 +100,16 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models: []
-shields: []
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models: []
+  shields: []
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/open-benchmark/run.yaml

@@ -142,109 +142,110 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: gpt-4o
-  provider_id: openai
-  provider_model_id: gpt-4o
-  model_type: llm
-- metadata: {}
-  model_id: claude-3-5-sonnet-latest
-  provider_id: anthropic
-  provider_model_id: claude-3-5-sonnet-latest
-  model_type: llm
-- metadata: {}
-  model_id: gemini/gemini-1.5-flash
-  provider_id: gemini
-  provider_model_id: gemini/gemini-1.5-flash
-  model_type: llm
-- metadata: {}
-  model_id: meta-llama/Llama-3.3-70B-Instruct
-  provider_id: groq
-  provider_model_id: groq/llama-3.3-70b-versatile
-  model_type: llm
-- metadata: {}
-  model_id: meta-llama/Llama-3.1-405B-Instruct
-  provider_id: together
-  provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
-  model_type: llm
-shields:
-- shield_id: meta-llama/Llama-Guard-3-8B
-vector_dbs: []
-datasets:
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/simpleqa?split=train
-  metadata: {}
-  dataset_id: simpleqa
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/mmlu_cot?split=test&name=all
-  metadata: {}
-  dataset_id: mmlu_cot
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main
-  metadata: {}
-  dataset_id: gpqa_cot
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/math_500?split=test
-  metadata: {}
-  dataset_id: math_500
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/IfEval?split=train
-  metadata: {}
-  dataset_id: ifeval
-- purpose: eval/messages-answer
-  source:
-    type: uri
-    uri: huggingface://datasets/llamastack/docvqa?split=val
-  metadata: {}
-  dataset_id: docvqa
-scoring_fns: []
-benchmarks:
-- dataset_id: simpleqa
-  scoring_functions:
-  - llm-as-judge::405b-simpleqa
-  metadata: {}
-  benchmark_id: meta-reference-simpleqa
-- dataset_id: mmlu_cot
-  scoring_functions:
-  - basic::regex_parser_multiple_choice_answer
-  metadata: {}
-  benchmark_id: meta-reference-mmlu-cot
-- dataset_id: gpqa_cot
-  scoring_functions:
-  - basic::regex_parser_multiple_choice_answer
-  metadata: {}
-  benchmark_id: meta-reference-gpqa-cot
-- dataset_id: math_500
-  scoring_functions:
-  - basic::regex_parser_math_response
-  metadata: {}
-  benchmark_id: meta-reference-math-500
-- dataset_id: ifeval
-  scoring_functions:
-  - basic::ifeval
-  metadata: {}
-  benchmark_id: meta-reference-ifeval
-- dataset_id: docvqa
-  scoring_functions:
-  - basic::docvqa
-  metadata: {}
-  benchmark_id: meta-reference-docvqa
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: gpt-4o
+    provider_id: openai
+    provider_model_id: gpt-4o
+    model_type: llm
+  - metadata: {}
+    model_id: claude-3-5-sonnet-latest
+    provider_id: anthropic
+    provider_model_id: claude-3-5-sonnet-latest
+    model_type: llm
+  - metadata: {}
+    model_id: gemini/gemini-1.5-flash
+    provider_id: gemini
+    provider_model_id: gemini/gemini-1.5-flash
+    model_type: llm
+  - metadata: {}
+    model_id: meta-llama/Llama-3.3-70B-Instruct
+    provider_id: groq
+    provider_model_id: groq/llama-3.3-70b-versatile
+    model_type: llm
+  - metadata: {}
+    model_id: meta-llama/Llama-3.1-405B-Instruct
+    provider_id: together
+    provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
+    model_type: llm
+  shields:
+  - shield_id: meta-llama/Llama-Guard-3-8B
+  vector_dbs: []
+  datasets:
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/simpleqa?split=train
+    metadata: {}
+    dataset_id: simpleqa
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/mmlu_cot?split=test&name=all
+    metadata: {}
+    dataset_id: mmlu_cot
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main
+    metadata: {}
+    dataset_id: gpqa_cot
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/math_500?split=test
+    metadata: {}
+    dataset_id: math_500
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/IfEval?split=train
+    metadata: {}
+    dataset_id: ifeval
+  - purpose: eval/messages-answer
+    source:
+      type: uri
+      uri: huggingface://datasets/llamastack/docvqa?split=val
+    metadata: {}
+    dataset_id: docvqa
+  scoring_fns: []
+  benchmarks:
+  - dataset_id: simpleqa
+    scoring_functions:
+    - llm-as-judge::405b-simpleqa
+    metadata: {}
+    benchmark_id: meta-reference-simpleqa
+  - dataset_id: mmlu_cot
+    scoring_functions:
+    - basic::regex_parser_multiple_choice_answer
+    metadata: {}
+    benchmark_id: meta-reference-mmlu-cot
+  - dataset_id: gpqa_cot
+    scoring_functions:
+    - basic::regex_parser_multiple_choice_answer
+    metadata: {}
+    benchmark_id: meta-reference-gpqa-cot
+  - dataset_id: math_500
+    scoring_functions:
+    - basic::regex_parser_math_response
+    metadata: {}
+    benchmark_id: meta-reference-math-500
+  - dataset_id: ifeval
+    scoring_functions:
+    - basic::ifeval
+    metadata: {}
+    benchmark_id: meta-reference-ifeval
+  - dataset_id: docvqa
+    scoring_functions:
+    - basic::docvqa
+    metadata: {}
+    benchmark_id: meta-reference-docvqa
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/postgres-demo/run.yaml

@@ -87,27 +87,28 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models:
-- metadata: {}
-  model_id: ${env.INFERENCE_MODEL}
-  provider_id: vllm-inference
-  model_type: llm
-- metadata:
-    embedding_dimension: 768
-  model_id: nomic-embed-text-v1.5
-  provider_id: sentence-transformers
-  model_type: embedding
-shields:
-- shield_id: meta-llama/Llama-Guard-3-8B
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models:
+  - metadata: {}
+    model_id: ${env.INFERENCE_MODEL}
+    provider_id: vllm-inference
+    model_type: llm
+  - metadata:
+      embedding_dimension: 768
+    model_id: nomic-embed-text-v1.5
+    provider_id: sentence-transformers
+    model_type: embedding
+  shields:
+  - shield_id: meta-llama/Llama-Guard-3-8B
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/starter-gpu/run.yaml

@@ -250,23 +250,24 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models: []
-shields:
-- shield_id: llama-guard
-  provider_id: ${env.SAFETY_MODEL:+llama-guard}
-  provider_shield_id: ${env.SAFETY_MODEL:=}
-- shield_id: code-scanner
-  provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
-  provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models: []
+  shields:
+  - shield_id: llama-guard
+    provider_id: ${env.SAFETY_MODEL:+llama-guard}
+    provider_shield_id: ${env.SAFETY_MODEL:=}
+  - shield_id: code-scanner
+    provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
+    provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/starter/run.yaml

@@ -247,23 +247,24 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models: []
-shields:
-- shield_id: llama-guard
-  provider_id: ${env.SAFETY_MODEL:+llama-guard}
-  provider_shield_id: ${env.SAFETY_MODEL:=}
-- shield_id: code-scanner
-  provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
-  provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models: []
+  shields:
+  - shield_id: llama-guard
+    provider_id: ${env.SAFETY_MODEL:+llama-guard}
+    provider_shield_id: ${env.SAFETY_MODEL:=}
+  - shield_id: code-scanner
+    provider_id: ${env.CODE_SCANNER_MODEL:+code-scanner}
+    provider_shield_id: ${env.CODE_SCANNER_MODEL:=}
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/template.py

@@ -272,13 +272,15 @@ class RunConfigSettings(BaseModel):
             "apis": apis,
             "providers": provider_configs,
             "storage": storage_config,
-            "models": [m.model_dump(exclude_none=True) for m in (self.default_models or [])],
-            "shields": [s.model_dump(exclude_none=True) for s in (self.default_shields or [])],
-            "vector_dbs": [],
-            "datasets": [d.model_dump(exclude_none=True) for d in (self.default_datasets or [])],
-            "scoring_fns": [],
-            "benchmarks": [b.model_dump(exclude_none=True) for b in (self.default_benchmarks or [])],
-            "tool_groups": [t.model_dump(exclude_none=True) for t in (self.default_tool_groups or [])],
+            "registered_resources": {
+                "models": [m.model_dump(exclude_none=True) for m in (self.default_models or [])],
+                "shields": [s.model_dump(exclude_none=True) for s in (self.default_shields or [])],
+                "vector_dbs": [],
+                "datasets": [d.model_dump(exclude_none=True) for d in (self.default_datasets or [])],
+                "scoring_fns": [],
+                "benchmarks": [b.model_dump(exclude_none=True) for b in (self.default_benchmarks or [])],
+                "tool_groups": [t.model_dump(exclude_none=True) for t in (self.default_tool_groups or [])],
+            },
             "server": {
                 "port": 8321,
             },

llama_stack/distributions/watsonx/run.yaml

@@ -115,17 +115,18 @@ storage:
     conversations:
       table_name: openai_conversations
       backend: sql_default
-models: []
-shields: []
-vector_dbs: []
-datasets: []
-scoring_fns: []
-benchmarks: []
-tool_groups:
-- toolgroup_id: builtin::websearch
-  provider_id: tavily-search
-- toolgroup_id: builtin::rag
-  provider_id: rag-runtime
+registered_resources:
+  models: []
+  shields: []
+  vector_dbs: []
+  datasets: []
+  scoring_fns: []
+  benchmarks: []
+  tool_groups:
+  - toolgroup_id: builtin::websearch
+    provider_id: tavily-search
+  - toolgroup_id: builtin::rag
+    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/providers/inline/telemetry/meta_reference/console_span_processor.py

@@ -1,75 +0,0 @@
-# Copyright (c) Meta Platforms, Inc. and affiliates.
-# All rights reserved.
-#
-# This source code is licensed under the terms described in the LICENSE file in
-# the root directory of this source tree.
-
-import json
-from datetime import UTC, datetime
-
-from opentelemetry.sdk.trace import ReadableSpan
-from opentelemetry.sdk.trace.export import SpanProcessor
-from opentelemetry.trace.status import StatusCode
-
-from llama_stack.log import get_logger
-
-logger = get_logger(name="console_span_processor", category="telemetry")
-
-
-class ConsoleSpanProcessor(SpanProcessor):
-    def __init__(self, print_attributes: bool = False):
-        self.print_attributes = print_attributes
-
-    def on_start(self, span: ReadableSpan, parent_context=None) -> None:
-        if span.attributes and span.attributes.get("__autotraced__"):
-            return
-
-        timestamp = datetime.fromtimestamp(span.start_time / 1e9, tz=UTC).strftime("%H:%M:%S.%f")[:-3]
-        logger.info(f"[dim]{timestamp}[/dim] [bold magenta][START][/bold magenta] [dim]{span.name}[/dim]")
-
-    def on_end(self, span: ReadableSpan) -> None:
-        timestamp = datetime.fromtimestamp(span.end_time / 1e9, tz=UTC).strftime("%H:%M:%S.%f")[:-3]
-        span_context = f"[dim]{timestamp}[/dim] [bold magenta][END][/bold magenta] [dim]{span.name}[/dim]"
-        if span.status.status_code == StatusCode.ERROR:
-            span_context += " [bold red][ERROR][/bold red]"
-        elif span.status.status_code != StatusCode.UNSET:
-            span_context += f" [{span.status.status_code}]"
-        duration_ms = (span.end_time - span.start_time) / 1e6
-        span_context += f" ({duration_ms:.2f}ms)"
-        logger.info(span_context)
-
-        if self.print_attributes and span.attributes:
-            for key, value in span.attributes.items():
-                if key.startswith("__"):
-                    continue
-                str_value = str(value)
-                if len(str_value) > 1000:
-                    str_value = str_value[:997] + "..."
-                logger.info(f"    [dim]{key}[/dim]: {str_value}")
-
-        for event in span.events:
-            event_time = datetime.fromtimestamp(event.timestamp / 1e9, tz=UTC).strftime("%H:%M:%S.%f")[:-3]
-            severity = event.attributes.get("severity", "info")
-            message = event.attributes.get("message", event.name)
-            if isinstance(message, dict) or isinstance(message, list):
-                message = json.dumps(message, indent=2)
-            severity_color = {
-                "error": "red",
-                "warn": "yellow",
-                "info": "white",
-                "debug": "dim",
-            }.get(severity, "white")
-            logger.info(f" {event_time} [bold {severity_color}][{severity.upper()}][/bold {severity_color}] {message}")
-            if event.attributes:
-                for key, value in event.attributes.items():
-                    if key.startswith("__") or key in ["message", "severity"]:
-                        continue
-                    logger.info(f"[dim]{key}[/dim]: {value}")
-
-    def shutdown(self) -> None:
-        """Shutdown the processor."""
-        pass
-
-    def force_flush(self, timeout_millis: float | None = None) -> bool:
-        """Force flush any pending spans."""
-        return True