feat: add list responses API (#2233)

# What does this PR do?
This is not part of the official OpenAI API, but we'll use this for the
logs UI.
In order to support more filtering options, I'm adopting the newly
introduced sql store in in place of the kv store.

## Test Plan
Added integration/unit tests.

llama_stack/apis/agents/agents.py

@@ -13,7 +13,7 @@ from typing import Annotated, Any, Literal, Protocol, runtime_checkable
 from pydantic import BaseModel, ConfigDict, Field
 
 from llama_stack.apis.common.content_types import URL, ContentDelta, InterleavedContent
-from llama_stack.apis.common.responses import PaginatedResponse
+from llama_stack.apis.common.responses import Order, PaginatedResponse
 from llama_stack.apis.inference import (
     CompletionMessage,
     ResponseFormat,
@@ -31,6 +31,7 @@ from llama_stack.apis.tools import ToolDef
 from llama_stack.schema_utils import json_schema_type, register_schema, webmethod
 
 from .openai_responses import (
+    ListOpenAIResponseObject,
     OpenAIResponseInput,
     OpenAIResponseInputTool,
     OpenAIResponseObject,
@@ -611,3 +612,21 @@ class Agents(Protocol):
         :returns: An OpenAIResponseObject.
         """
         ...
+
+    @webmethod(route="/openai/v1/responses", method="GET")
+    async def list_openai_responses(
+        self,
+        after: str | None = None,
+        limit: int | None = 50,
+        model: str | None = None,
+        order: Order | None = Order.desc,
+    ) -> ListOpenAIResponseObject:
+        """List all OpenAI responses.
+
+        :param after: The ID of the last response to return.
+        :param limit: The number of responses to return.
+        :param model: The model to filter responses by.
+        :param order: The order to sort responses by when sorted by created_at ('asc' or 'desc').
+        :returns: A ListOpenAIResponseObject.
+        """
+        ...

llama_stack/apis/agents/openai_responses.py

@@ -219,3 +219,17 @@ register_schema(OpenAIResponseInputTool, name="OpenAIResponseInputTool")
 class OpenAIResponseInputItemList(BaseModel):
     data: list[OpenAIResponseInput]
     object: Literal["list"] = "list"
+
+
+@json_schema_type
+class OpenAIResponseObjectWithInput(OpenAIResponseObject):
+    input: list[OpenAIResponseInput]
+
+
+@json_schema_type
+class ListOpenAIResponseObject(BaseModel):
+    data: list[OpenAIResponseObjectWithInput]
+    has_more: bool
+    first_id: str
+    last_id: str
+    object: Literal["list"] = "list"

llama_stack/apis/common/responses.py

@@ -4,6 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
+from enum import Enum
 from typing import Any
 
 from pydantic import BaseModel
@@ -11,6 +12,11 @@ from pydantic import BaseModel
 from llama_stack.schema_utils import json_schema_type
 
 
+class Order(Enum):
+    asc = "asc"
+    desc = "desc"
+
+
 @json_schema_type
 class PaginatedResponse(BaseModel):
     """A generic paginated response that follows a simple format.

llama_stack/apis/inference/inference.py

@@ -19,6 +19,7 @@ from pydantic import BaseModel, Field, field_validator
 from typing_extensions import TypedDict
 
 from llama_stack.apis.common.content_types import ContentDelta, InterleavedContent, InterleavedContentItem
+from llama_stack.apis.common.responses import Order
 from llama_stack.apis.models import Model
 from llama_stack.apis.telemetry.telemetry import MetricResponseMixin
 from llama_stack.models.llama.datatypes import (
@@ -833,11 +834,6 @@ class ListOpenAIChatCompletionResponse(BaseModel):
     object: Literal["list"] = "list"
 
 
-class Order(Enum):
-    asc = "asc"
-    desc = "desc"
-
-
 @runtime_checkable
 @trace_protocol
 class InferenceProvider(Protocol):

llama_stack/providers/inline/agents/meta_reference/agents.py

@@ -20,9 +20,11 @@ from llama_stack.apis.agents import (
     AgentTurnCreateRequest,
     AgentTurnResumeRequest,
     Document,
+    ListOpenAIResponseObject,
     OpenAIResponseInput,
     OpenAIResponseInputTool,
     OpenAIResponseObject,
+    Order,
     Session,
     Turn,
 )
@@ -39,6 +41,7 @@ from llama_stack.apis.tools import ToolGroups, ToolRuntime
 from llama_stack.apis.vector_io import VectorIO
 from llama_stack.providers.utils.kvstore import InmemoryKVStoreImpl, kvstore_impl
 from llama_stack.providers.utils.pagination import paginate_records
+from llama_stack.providers.utils.responses.responses_store import ResponsesStore
 
 from .agent_instance import ChatAgent
 from .config import MetaReferenceAgentsImplConfig
@@ -66,15 +69,17 @@ class MetaReferenceAgentsImpl(Agents):
         self.tool_groups_api = tool_groups_api
 
         self.in_memory_store = InmemoryKVStoreImpl()
-        self.openai_responses_impl = None
+        self.openai_responses_impl: OpenAIResponsesImpl | None = None
 
     async def initialize(self) -> None:
         self.persistence_store = await kvstore_impl(self.config.persistence_store)
+        self.responses_store = ResponsesStore(self.config.responses_store)
+        await self.responses_store.initialize()
         self.openai_responses_impl = OpenAIResponsesImpl(
-            self.persistence_store,
             inference_api=self.inference_api,
             tool_groups_api=self.tool_groups_api,
             tool_runtime_api=self.tool_runtime_api,
+            responses_store=self.responses_store,
         )
 
     async def create_agent(
@@ -323,3 +328,12 @@ class MetaReferenceAgentsImpl(Agents):
         return await self.openai_responses_impl.create_openai_response(
             input, model, instructions, previous_response_id, store, stream, temperature, tools
         )
+
+    async def list_openai_responses(
+        self,
+        after: str | None = None,
+        limit: int | None = 50,
+        model: str | None = None,
+        order: Order | None = Order.desc,
+    ) -> ListOpenAIResponseObject:
+        return await self.openai_responses_impl.list_openai_responses(after, limit, model, order)

llama_stack/providers/inline/agents/meta_reference/config.py

@@ -10,10 +10,12 @@ from pydantic import BaseModel
 
 from llama_stack.providers.utils.kvstore import KVStoreConfig
 from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
+from llama_stack.providers.utils.sqlstore.sqlstore import SqliteSqlStoreConfig, SqlStoreConfig
 
 
 class MetaReferenceAgentsImplConfig(BaseModel):
     persistence_store: KVStoreConfig
+    responses_store: SqlStoreConfig
 
     @classmethod
     def sample_run_config(cls, __distro_dir__: str) -> dict[str, Any]:
@@ -21,5 +23,9 @@ class MetaReferenceAgentsImplConfig(BaseModel):
             "persistence_store": SqliteKVStoreConfig.sample_run_config(
                 __distro_dir__=__distro_dir__,
                 db_name="agents_store.db",
-            )
+            ),
+            "responses_store": SqliteSqlStoreConfig.sample_run_config(
+                __distro_dir__=__distro_dir__,
+                db_name="responses_store.db",
+            ),
         }

llama_stack/providers/inline/agents/meta_reference/openai_responses.py

@@ -12,7 +12,9 @@ from typing import Any, cast
 from openai.types.chat import ChatCompletionToolParam
 from pydantic import BaseModel
 
+from llama_stack.apis.agents import Order
 from llama_stack.apis.agents.openai_responses import (
+    ListOpenAIResponseObject,
     OpenAIResponseInput,
     OpenAIResponseInputFunctionToolCallOutput,
     OpenAIResponseInputItemList,
@@ -53,7 +55,7 @@ from llama_stack.apis.tools.tools import ToolGroups, ToolInvocationResult, ToolR
 from llama_stack.log import get_logger
 from llama_stack.models.llama.datatypes import ToolDefinition, ToolParamDefinition
 from llama_stack.providers.utils.inference.openai_compat import convert_tooldef_to_openai_tool
-from llama_stack.providers.utils.kvstore import KVStore
+from llama_stack.providers.utils.responses.responses_store import ResponsesStore
 
 logger = get_logger(name=__name__, category="openai_responses")
 
@@ -169,34 +171,27 @@ class OpenAIResponsePreviousResponseWithInputItems(BaseModel):
 class OpenAIResponsesImpl:
     def __init__(
         self,
-        persistence_store: KVStore,
         inference_api: Inference,
         tool_groups_api: ToolGroups,
         tool_runtime_api: ToolRuntime,
+        responses_store: ResponsesStore,
     ):
-        self.persistence_store = persistence_store
         self.inference_api = inference_api
         self.tool_groups_api = tool_groups_api
         self.tool_runtime_api = tool_runtime_api
-
-    async def _get_previous_response_with_input(self, id: str) -> OpenAIResponsePreviousResponseWithInputItems:
-        key = f"{OPENAI_RESPONSES_PREFIX}{id}"
-        response_json = await self.persistence_store.get(key=key)
-        if response_json is None:
-            raise ValueError(f"OpenAI response with id '{id}' not found")
-        return OpenAIResponsePreviousResponseWithInputItems.model_validate_json(response_json)
+        self.responses_store = responses_store
 
     async def _prepend_previous_response(
         self, input: str | list[OpenAIResponseInput], previous_response_id: str | None = None
     ):
         if previous_response_id:
-            previous_response_with_input = await self._get_previous_response_with_input(previous_response_id)
+            previous_response_with_input = await self.responses_store.get_response_object(previous_response_id)
 
             # previous response input items
-            new_input_items = previous_response_with_input.input_items.data
+            new_input_items = previous_response_with_input.input
 
             # previous response output items
-            new_input_items.extend(previous_response_with_input.response.output)
+            new_input_items.extend(previous_response_with_input.output)
 
             # new input items from the current request
             if isinstance(input, str):
@@ -216,8 +211,17 @@ class OpenAIResponsesImpl:
         self,
         response_id: str,
     ) -> OpenAIResponseObject:
-        response_with_input = await self._get_previous_response_with_input(response_id)
-        return response_with_input.response
+        response_with_input = await self.responses_store.get_response_object(response_id)
+        return OpenAIResponseObject(**{k: v for k, v in response_with_input.model_dump().items() if k != "input"})
+
+    async def list_openai_responses(
+        self,
+        after: str | None = None,
+        limit: int | None = 50,
+        model: str | None = None,
+        order: Order | None = Order.desc,
+    ) -> ListOpenAIResponseObject:
+        return await self.responses_store.list_responses(after, limit, model, order)
 
     async def create_openai_response(
         self,
@@ -360,15 +364,9 @@ class OpenAIResponsesImpl:
                     else:
                         input_items_data.append(input_item)
 
-            input_items = OpenAIResponseInputItemList(data=input_items_data)
-            prev_response = OpenAIResponsePreviousResponseWithInputItems(
-                input_items=input_items,
-                response=response,
-            )
-            key = f"{OPENAI_RESPONSES_PREFIX}{response.id}"
-            await self.persistence_store.set(
-                key=key,
-                value=prev_response.model_dump_json(),
+            await self.responses_store.store_response_object(
+                response_object=response,
+                input=input_items_data,
             )
 
         if stream:

llama_stack/providers/utils/responses/responses_store.py

@@ -0,0 +1,98 @@
+# 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.
+from llama_stack.apis.agents import (
+    Order,
+)
+from llama_stack.apis.agents.openai_responses import (
+    ListOpenAIResponseObject,
+    OpenAIResponseInput,
+    OpenAIResponseObject,
+    OpenAIResponseObjectWithInput,
+)
+from llama_stack.distribution.utils.config_dirs import RUNTIME_BASE_DIR
+
+from ..sqlstore.api import ColumnDefinition, ColumnType
+from ..sqlstore.sqlstore import SqliteSqlStoreConfig, SqlStoreConfig, sqlstore_impl
+
+
+class ResponsesStore:
+    def __init__(self, sql_store_config: SqlStoreConfig):
+        if not sql_store_config:
+            sql_store_config = SqliteSqlStoreConfig(
+                db_path=(RUNTIME_BASE_DIR / "sqlstore.db").as_posix(),
+            )
+        self.sql_store = sqlstore_impl(sql_store_config)
+
+    async def initialize(self):
+        """Create the necessary tables if they don't exist."""
+        await self.sql_store.create_table(
+            "openai_responses",
+            {
+                "id": ColumnDefinition(type=ColumnType.STRING, primary_key=True),
+                "created_at": ColumnType.INTEGER,
+                "response_object": ColumnType.JSON,
+                "model": ColumnType.STRING,
+            },
+        )
+
+    async def store_response_object(
+        self, response_object: OpenAIResponseObject, input: list[OpenAIResponseInput]
+    ) -> None:
+        data = response_object.model_dump()
+        data["input"] = [input_item.model_dump() for input_item in input]
+
+        await self.sql_store.insert(
+            "openai_responses",
+            {
+                "id": data["id"],
+                "created_at": data["created_at"],
+                "model": data["model"],
+                "response_object": data,
+            },
+        )
+
+    async def list_responses(
+        self,
+        after: str | None = None,
+        limit: int | None = 50,
+        model: str | None = None,
+        order: Order | None = Order.desc,
+    ) -> ListOpenAIResponseObject:
+        """
+        List responses from the database.
+
+        :param after: The ID of the last response to return.
+        :param limit: The maximum number of responses to return.
+        :param model: The model to filter by.
+        :param order: The order to sort the responses by.
+        """
+        # TODO: support after
+        if after:
+            raise NotImplementedError("After is not supported for SQLite")
+        if not order:
+            order = Order.desc
+
+        rows = await self.sql_store.fetch_all(
+            "openai_responses",
+            where={"model": model} if model else None,
+            order_by=[("created_at", order.value)],
+            limit=limit,
+        )
+
+        data = [OpenAIResponseObjectWithInput(**row["response_object"]) for row in rows]
+        return ListOpenAIResponseObject(
+            data=data,
+            # TODO: implement has_more
+            has_more=False,
+            first_id=data[0].id if data else "",
+            last_id=data[-1].id if data else "",
+        )
+
+    async def get_response_object(self, response_id: str) -> OpenAIResponseObjectWithInput:
+        row = await self.sql_store.fetch_one("openai_responses", where={"id": response_id})
+        if not row:
+            raise ValueError(f"Response with id {response_id} not found") from None
+        return OpenAIResponseObjectWithInput(**row["response_object"])

llama_stack/templates/bedrock/run.yaml

@@ -35,6 +35,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/bedrock}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/bedrock}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/cerebras/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/cerebras}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/cerebras}/responses_store.db
   eval:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/ci-tests/run.yaml

@@ -38,6 +38,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ci-tests}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ci-tests}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

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

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/dell}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/dell}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/dell/run.yaml

@@ -37,6 +37,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/dell}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/dell}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/fireworks/run-with-safety.yaml

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/fireworks/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/groq/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/hf-endpoint/run-with-safety.yaml

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/hf-endpoint/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-endpoint}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/hf-serverless/run-with-safety.yaml

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/hf-serverless/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/hf-serverless}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/llama_api/run.yaml

@@ -50,6 +50,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/llama_api}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/llama_api}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

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

@@ -56,6 +56,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

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

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

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

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/nvidia/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/ollama/run-with-safety.yaml

@@ -40,6 +40,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/ollama/run.yaml

@@ -38,6 +38,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ollama}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/open-benchmark/run.yaml

@@ -64,6 +64,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/passthrough/run-with-safety.yaml

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/passthrough}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/passthrough}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/passthrough/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/passthrough}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/passthrough}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/remote-vllm/run-with-safety.yaml

@@ -50,6 +50,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/responses_store.db
   eval:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/remote-vllm/run.yaml

@@ -43,6 +43,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/remote-vllm}/responses_store.db
   eval:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/sambanova/run.yaml

@@ -51,6 +51,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/sambanova}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/sambanova}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/starter/run.yaml

@@ -72,6 +72,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/starter}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/starter}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/tgi/run-with-safety.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/tgi/run.yaml

@@ -40,6 +40,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/tgi}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/together/run-with-safety.yaml

@@ -46,6 +46,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/together/run.yaml

@@ -41,6 +41,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/verification/run.yaml

@@ -74,6 +74,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/verification}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/verification}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/vllm-gpu/run.yaml

@@ -45,6 +45,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/vllm-gpu}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/vllm-gpu}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference

llama_stack/templates/watsonx/run.yaml

@@ -42,6 +42,9 @@ providers:
         type: sqlite
         namespace: null
         db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/watsonx}/agents_store.db
+      responses_store:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/watsonx}/responses_store.db
   telemetry:
   - provider_id: meta-reference
     provider_type: inline::meta-reference