refactor: enforce top-level imports for llama-stack-api

Enforce that all imports from llama-stack-api use the form:

from llama_stack_api import <symbol>

 This prevents external code from accessing internal package structure
 (e.g., llama_stack_api.agents, llama_stack_api.common.*) and establishes
 a clear public API boundary.

 Changes:
 - Export 400+ symbols from llama_stack_api/__init__.py
 - Include all API types, common utilities, and strong_typing helpers
 - Update files across src/llama_stack, docs/, tests/, scripts/
 - Convert all submodule imports to top-level imports
 - ensure docs use the proper importing structure

 Addresses PR review feedback requiring explicit __all__ definition to
 prevent "peeking inside" the API package.

Signed-off-by: Charlie Doern <cdoern@redhat.com>

.github/workflows/python-build-test.yml

@@ -30,14 +30,14 @@ jobs:
         activate-environment: true
         version: 0.7.6
 
-    - name: Build Llama Stack Spec package
+    - name: Build Llama Stack API package
       working-directory: src/llama-stack-api
       run: uv build
 
     - name: Build Llama Stack package
       run: uv build
 
-    - name: Install Llama Stack package (with spec from local build)
+    - name: Install Llama Stack package (with api stubs from local build)
       run: |
         uv pip install --find-links src/llama-stack-api/dist dist/*.whl
 

docs/docs/concepts/apis/external.mdx

@@ -58,7 +58,7 @@ External APIs must expose a `available_providers()` function in their module tha
 
 ```python
 # llama_stack_api_weather/api.py
-from llama_stack_api.providers.datatypes import Api, InlineProviderSpec, ProviderSpec
+from llama_stack_api import Api, InlineProviderSpec, ProviderSpec
 
 
 def available_providers() -> list[ProviderSpec]:
@@ -79,7 +79,7 @@ A Protocol class like so:
 # llama_stack_api_weather/api.py
 from typing import Protocol
 
-from llama_stack_api.schema_utils import webmethod
+from llama_stack_api import webmethod
 
 
 class WeatherAPI(Protocol):
@@ -151,13 +151,12 @@ __all__ = ["WeatherAPI", "available_providers"]
 # llama-stack-api-weather/src/llama_stack_api_weather/weather.py
 from typing import Protocol
 
-from llama_stack_api.providers.datatypes import (
+from llama_stack_api import (
     Api,
     ProviderSpec,
     RemoteProviderSpec,
+    webmethod,
 )
-from llama_stack_api.schema_utils import webmethod
-
 
 def available_providers() -> list[ProviderSpec]:
     return [

docs/docs/providers/vector_io/inline_sqlite-vec.mdx

@@ -153,7 +153,7 @@ description: |
   Example using RAGQueryConfig with different search modes:
 
   ```python
-  from llama_stack_api.rag_tool import RAGQueryConfig, RRFRanker, WeightedRanker
+  from llama_stack_api import RAGQueryConfig, RRFRanker, WeightedRanker
 
   # Vector search
   config = RAGQueryConfig(mode="vector", max_chunks=5)
@@ -358,7 +358,7 @@ Two ranker types are supported:
 Example using RAGQueryConfig with different search modes:
 
 ```python
-from llama_stack_api.rag_tool import RAGQueryConfig, RRFRanker, WeightedRanker
+from llama_stack_api import RAGQueryConfig, RRFRanker, WeightedRanker
 
 # Vector search
 config = RAGQueryConfig(mode="vector", max_chunks=5)

docs/openapi_generator/generate.py

@@ -16,7 +16,7 @@ import sys
 import fire
 import ruamel.yaml as yaml
 
-from llama_stack_api.version import LLAMA_STACK_API_V1 # noqa: E402
+from llama_stack_api import LLAMA_STACK_API_V1 # noqa: E402
 from llama_stack.core.stack import LlamaStack  # noqa: E402
 
 from .pyopenapi.options import Options  # noqa: E402

docs/openapi_generator/pyopenapi/generator.py

@@ -16,27 +16,27 @@ from typing import Annotated, Any, Dict, get_args, get_origin, Set, Union
 
 from fastapi import UploadFile
 
-from llama_stack_api.datatypes import Error
+from llama_stack_api import (
-from llama_stack_api.strong_typing.core import JsonType
+    Docstring,
-from llama_stack_api.strong_typing.docstring import Docstring, parse_type
+    Error,
-from llama_stack_api.strong_typing.inspection import (
+    JsonSchemaGenerator,
+    JsonType,
+    Schema,
+    SchemaOptions,
+    get_schema_identifier,
     is_generic_list,
     is_type_optional,
     is_type_union,
     is_unwrapped_body_param,
+    json_dump_string,
+    object_to_json,
+    parse_type,
+    python_type_to_name,
+    register_schema,
     unwrap_generic_list,
     unwrap_optional_type,
     unwrap_union_types,
 )
-from llama_stack_api.strong_typing.name import python_type_to_name
-from llama_stack_api.strong_typing.schema import (
-    get_schema_identifier,
-    JsonSchemaGenerator,
-    register_schema,
-    Schema,
-    SchemaOptions,
-)
-from llama_stack_api.strong_typing.serialization import json_dump_string, object_to_json
 from pydantic import BaseModel
 
 from .operations import (

docs/openapi_generator/pyopenapi/operations.py

@@ -11,19 +11,21 @@ import typing
 from dataclasses import dataclass
 from typing import Any, Callable, Dict, Iterable, Iterator, List, Optional, Tuple, Union
 
-from llama_stack_api.version import LLAMA_STACK_API_V1, LLAMA_STACK_API_V1BETA, LLAMA_STACK_API_V1ALPHA
-
 from termcolor import colored
 
-from llama_stack_api.strong_typing.inspection import get_signature
-
 from typing import get_origin, get_args
 
 from fastapi import UploadFile
 from fastapi.params import File, Form
 from typing import Annotated
 
-from llama_stack_api.schema_utils import ExtraBodyField
+from llama_stack_api import (
+    ExtraBodyField,
+    LLAMA_STACK_API_V1,
+    LLAMA_STACK_API_V1ALPHA,
+    LLAMA_STACK_API_V1BETA,
+    get_signature,
+)
 
 
 def split_prefix(

docs/openapi_generator/pyopenapi/specification.py

@@ -9,7 +9,7 @@ import enum
 from dataclasses import dataclass
 from typing import Any, ClassVar, Dict, List, Optional, Union
 
-from llama_stack_api.strong_typing.schema import JsonType, Schema, StrictJsonType
+from llama_stack_api import JsonType, Schema, StrictJsonType
 
 URL = str
 

docs/openapi_generator/pyopenapi/utility.py

@@ -11,8 +11,7 @@ from pathlib import Path
 from typing import Any, List, Optional, TextIO, Union, get_type_hints, get_origin, get_args
 
 from pydantic import BaseModel
-from llama_stack_api.strong_typing.schema import object_to_json, StrictJsonType
+from llama_stack_api import StrictJsonType, is_unwrapped_body_param, object_to_json
-from llama_stack_api.strong_typing.inspection import is_unwrapped_body_param
 from llama_stack.core.resolver import api_protocol_map
 
 from .generator import Generator
@@ -165,12 +164,12 @@ def _validate_api_delete_method_returns_none(method) -> str | None:
         return "has no return type annotation"
 
     return_type = hints['return']
-    
+
     # Allow OpenAI endpoints to return response objects since they follow OpenAI specification
     method_name = getattr(method, '__name__', '')
     if method_name.__contains__('openai_'):
         return None
-    
+
     if return_type is not None and return_type is not type(None):
         return "does not return None where None is mandatory"
 

scripts/generate_prompt_format.py

@@ -14,7 +14,7 @@ import os
 from pathlib import Path
 
 import fire
-from llama_stack_api.common.errors import ModelNotFoundError
+from llama_stack_api import ModelNotFoundError
 
 from llama_stack.models.llama.llama3.generation import Llama3
 from llama_stack.models.llama.llama4.generation import Llama4

src/llama-stack-api/llama_stack_api/__init__.py

@@ -12,16 +12,860 @@ for Llama Stack. It is designed to be a lightweight dependency for external prov
 and clients that need to interact with Llama Stack APIs without requiring the full
 server implementation.
 
-Key components:
+All imports from this package MUST use the form:
-- API modules (agents, inference, safety, etc.): Protocol definitions for all Llama Stack APIs
+    from llama_stack_api import <symbol>
-- datatypes: Core data types and provider specifications
+
-- common: Common data types used across APIs
+Sub-module imports (e.g., from llama_stack_api.agents import Agents) are NOT supported
-- strong_typing: Type system utilities
+and considered a code smell. All exported symbols are explicitly listed in __all__.
-- schema_utils: Schema validation and utilities
 """
 
-__version__ = "0.1.0"
+__version__ = "0.4.0"
 
-from . import common, datatypes, schema_utils, strong_typing  # noqa: F401
+# Import submodules for those who need them
+from . import common, strong_typing  # noqa: F401
 
-__all__ = ["common", "datatypes", "schema_utils", "strong_typing"]
+# Import all public API symbols
+from .agents import Agents, ResponseGuardrail, ResponseGuardrailSpec
+from .batches import Batches, BatchObject, ListBatchesResponse
+from .benchmarks import (
+    Benchmark,
+    BenchmarkInput,
+    Benchmarks,
+    CommonBenchmarkFields,
+    ListBenchmarksResponse,
+)
+
+# Import commonly used types from common submodule
+from .common.content_types import (
+    URL,
+    ImageContentItem,
+    InterleavedContent,
+    InterleavedContentItem,
+    TextContentItem,
+    _URLOrData,
+)
+from .common.errors import (
+    ConflictError,
+    DatasetNotFoundError,
+    InvalidConversationIdError,
+    ModelNotFoundError,
+    ModelTypeError,
+    ResourceNotFoundError,
+    TokenValidationError,
+    ToolGroupNotFoundError,
+    UnsupportedModelError,
+    VectorStoreNotFoundError,
+)
+from .common.job_types import Job, JobStatus
+from .common.responses import Order, PaginatedResponse
+from .common.training_types import Checkpoint, PostTrainingMetric
+from .common.type_system import (
+    ChatCompletionInputType,
+    CompletionInputType,
+    NumberType,
+    ParamType,
+    StringType,
+)
+from .conversations import (
+    Conversation,
+    ConversationDeletedResource,
+    ConversationItem,
+    ConversationItemCreateRequest,
+    ConversationItemDeletedResource,
+    ConversationItemInclude,
+    ConversationItemList,
+    ConversationMessage,
+    Conversations,
+    Metadata,
+)
+from .datasetio import DatasetIO, DatasetStore
+from .datasets import (
+    CommonDatasetFields,
+    Dataset,
+    DatasetInput,
+    DatasetPurpose,
+    Datasets,
+    DatasetType,
+    DataSource,
+    ListDatasetsResponse,
+    RowsDataSource,
+    URIDataSource,
+)
+from .datatypes import (
+    Api,
+    BenchmarksProtocolPrivate,
+    DatasetsProtocolPrivate,
+    DynamicApiMeta,
+    Error,
+    ExternalApiSpec,
+    HealthResponse,
+    HealthStatus,
+    InlineProviderSpec,
+    ModelsProtocolPrivate,
+    ProviderSpec,
+    RemoteProviderConfig,
+    RemoteProviderSpec,
+    RoutingTable,
+    ScoringFunctionsProtocolPrivate,
+    ShieldsProtocolPrivate,
+    ToolGroupsProtocolPrivate,
+    VectorStoresProtocolPrivate,
+)
+from .eval import BenchmarkConfig, Eval, EvalCandidate, EvaluateResponse, ModelCandidate
+from .files import (
+    ExpiresAfter,
+    Files,
+    ListOpenAIFileResponse,
+    OpenAIFileDeleteResponse,
+    OpenAIFileObject,
+    OpenAIFilePurpose,
+)
+from .inference import (
+    Bf16QuantizationConfig,
+    ChatCompletionResponseEventType,
+    CompletionRequest,
+    EmbeddingsResponse,
+    EmbeddingTaskType,
+    Fp8QuantizationConfig,
+    GrammarResponseFormat,
+    GreedySamplingStrategy,
+    Inference,
+    InferenceProvider,
+    Int4QuantizationConfig,
+    JsonSchemaResponseFormat,
+    ListOpenAIChatCompletionResponse,
+    LogProbConfig,
+    ModelStore,
+    OpenAIAssistantMessageParam,
+    OpenAIChatCompletion,
+    OpenAIChatCompletionChunk,
+    OpenAIChatCompletionContentPartImageParam,
+    OpenAIChatCompletionContentPartParam,
+    OpenAIChatCompletionContentPartTextParam,
+    OpenAIChatCompletionMessageContent,
+    OpenAIChatCompletionRequestWithExtraBody,
+    OpenAIChatCompletionTextOnlyMessageContent,
+    OpenAIChatCompletionToolCall,
+    OpenAIChatCompletionToolCallFunction,
+    OpenAIChatCompletionUsage,
+    OpenAIChatCompletionUsageCompletionTokensDetails,
+    OpenAIChatCompletionUsagePromptTokensDetails,
+    OpenAIChoice,
+    OpenAIChoiceDelta,
+    OpenAIChoiceLogprobs,
+    OpenAIChunkChoice,
+    OpenAICompletion,
+    OpenAICompletionChoice,
+    OpenAICompletionLogprobs,
+    OpenAICompletionRequestWithExtraBody,
+    OpenAICompletionWithInputMessages,
+    OpenAIDeveloperMessageParam,
+    OpenAIEmbeddingData,
+    OpenAIEmbeddingsRequestWithExtraBody,
+    OpenAIEmbeddingsResponse,
+    OpenAIEmbeddingUsage,
+    OpenAIFile,
+    OpenAIFileFile,
+    OpenAIImageURL,
+    OpenAIJSONSchema,
+    OpenAIMessageParam,
+    OpenAIResponseFormatJSONObject,
+    OpenAIResponseFormatJSONSchema,
+    OpenAIResponseFormatParam,
+    OpenAIResponseFormatText,
+    OpenAISystemMessageParam,
+    OpenAITokenLogProb,
+    OpenAIToolMessageParam,
+    OpenAITopLogProb,
+    OpenAIUserMessageParam,
+    QuantizationConfig,
+    QuantizationType,
+    RerankData,
+    RerankResponse,
+    ResponseFormat,
+    ResponseFormatType,
+    SamplingParams,
+    SamplingStrategy,
+    SystemMessage,
+    SystemMessageBehavior,
+    TextTruncation,
+    TokenLogProbs,
+    ToolChoice,
+    ToolResponseMessage,
+    TopKSamplingStrategy,
+    TopPSamplingStrategy,
+    UserMessage,
+)
+from .inspect import (
+    ApiFilter,
+    HealthInfo,
+    Inspect,
+    ListRoutesResponse,
+    RouteInfo,
+    VersionInfo,
+)
+from .models import (
+    CommonModelFields,
+    ListModelsResponse,
+    Model,
+    ModelInput,
+    Models,
+    ModelType,
+    OpenAIListModelsResponse,
+    OpenAIModel,
+)
+from .openai_responses import (
+    AllowedToolsFilter,
+    ApprovalFilter,
+    ListOpenAIResponseInputItem,
+    ListOpenAIResponseObject,
+    MCPListToolsTool,
+    OpenAIDeleteResponseObject,
+    OpenAIResponseAnnotationCitation,
+    OpenAIResponseAnnotationContainerFileCitation,
+    OpenAIResponseAnnotationFileCitation,
+    OpenAIResponseAnnotationFilePath,
+    OpenAIResponseAnnotations,
+    OpenAIResponseContentPart,
+    OpenAIResponseContentPartOutputText,
+    OpenAIResponseContentPartReasoningSummary,
+    OpenAIResponseContentPartReasoningText,
+    OpenAIResponseContentPartRefusal,
+    OpenAIResponseError,
+    OpenAIResponseInput,
+    OpenAIResponseInputFunctionToolCallOutput,
+    OpenAIResponseInputMessageContent,
+    OpenAIResponseInputMessageContentFile,
+    OpenAIResponseInputMessageContentImage,
+    OpenAIResponseInputMessageContentText,
+    OpenAIResponseInputTool,
+    OpenAIResponseInputToolFileSearch,
+    OpenAIResponseInputToolFunction,
+    OpenAIResponseInputToolMCP,
+    OpenAIResponseInputToolWebSearch,
+    OpenAIResponseMCPApprovalRequest,
+    OpenAIResponseMCPApprovalResponse,
+    OpenAIResponseMessage,
+    OpenAIResponseObject,
+    OpenAIResponseObjectStream,
+    OpenAIResponseObjectStreamResponseCompleted,
+    OpenAIResponseObjectStreamResponseContentPartAdded,
+    OpenAIResponseObjectStreamResponseContentPartDone,
+    OpenAIResponseObjectStreamResponseCreated,
+    OpenAIResponseObjectStreamResponseFailed,
+    OpenAIResponseObjectStreamResponseFileSearchCallCompleted,
+    OpenAIResponseObjectStreamResponseFileSearchCallInProgress,
+    OpenAIResponseObjectStreamResponseFileSearchCallSearching,
+    OpenAIResponseObjectStreamResponseFunctionCallArgumentsDelta,
+    OpenAIResponseObjectStreamResponseFunctionCallArgumentsDone,
+    OpenAIResponseObjectStreamResponseIncomplete,
+    OpenAIResponseObjectStreamResponseInProgress,
+    OpenAIResponseObjectStreamResponseMcpCallArgumentsDelta,
+    OpenAIResponseObjectStreamResponseMcpCallArgumentsDone,
+    OpenAIResponseObjectStreamResponseMcpCallCompleted,
+    OpenAIResponseObjectStreamResponseMcpCallFailed,
+    OpenAIResponseObjectStreamResponseMcpCallInProgress,
+    OpenAIResponseObjectStreamResponseMcpListToolsCompleted,
+    OpenAIResponseObjectStreamResponseMcpListToolsFailed,
+    OpenAIResponseObjectStreamResponseMcpListToolsInProgress,
+    OpenAIResponseObjectStreamResponseOutputItemAdded,
+    OpenAIResponseObjectStreamResponseOutputItemDone,
+    OpenAIResponseObjectStreamResponseOutputTextAnnotationAdded,
+    OpenAIResponseObjectStreamResponseOutputTextDelta,
+    OpenAIResponseObjectStreamResponseOutputTextDone,
+    OpenAIResponseObjectStreamResponseReasoningSummaryPartAdded,
+    OpenAIResponseObjectStreamResponseReasoningSummaryPartDone,
+    OpenAIResponseObjectStreamResponseReasoningSummaryTextDelta,
+    OpenAIResponseObjectStreamResponseReasoningSummaryTextDone,
+    OpenAIResponseObjectStreamResponseReasoningTextDelta,
+    OpenAIResponseObjectStreamResponseReasoningTextDone,
+    OpenAIResponseObjectStreamResponseRefusalDelta,
+    OpenAIResponseObjectStreamResponseRefusalDone,
+    OpenAIResponseObjectStreamResponseWebSearchCallCompleted,
+    OpenAIResponseObjectStreamResponseWebSearchCallInProgress,
+    OpenAIResponseObjectStreamResponseWebSearchCallSearching,
+    OpenAIResponseObjectWithInput,
+    OpenAIResponseOutput,
+    OpenAIResponseOutputMessageContent,
+    OpenAIResponseOutputMessageContentOutputText,
+    OpenAIResponseOutputMessageFileSearchToolCall,
+    OpenAIResponseOutputMessageFileSearchToolCallResults,
+    OpenAIResponseOutputMessageFunctionToolCall,
+    OpenAIResponseOutputMessageMCPCall,
+    OpenAIResponseOutputMessageMCPListTools,
+    OpenAIResponseOutputMessageWebSearchToolCall,
+    OpenAIResponsePrompt,
+    OpenAIResponseText,
+    OpenAIResponseTextFormat,
+    OpenAIResponseTool,
+    OpenAIResponseToolMCP,
+    OpenAIResponseUsage,
+    OpenAIResponseUsageInputTokensDetails,
+    OpenAIResponseUsageOutputTokensDetails,
+    WebSearchToolTypes,
+)
+from .post_training import (
+    AlgorithmConfig,
+    DataConfig,
+    DatasetFormat,
+    DPOAlignmentConfig,
+    DPOLossType,
+    EfficiencyConfig,
+    ListPostTrainingJobsResponse,
+    LoraFinetuningConfig,
+    OptimizerConfig,
+    OptimizerType,
+    PostTraining,
+    PostTrainingJob,
+    PostTrainingJobArtifactsResponse,
+    PostTrainingJobLogStream,
+    PostTrainingJobStatusResponse,
+    PostTrainingRLHFRequest,
+    QATFinetuningConfig,
+    RLHFAlgorithm,
+    TrainingConfig,
+)
+from .prompts import ListPromptsResponse, Prompt, Prompts
+from .providers import ListProvidersResponse, ProviderInfo, Providers
+from .rag_tool import (
+    DefaultRAGQueryGeneratorConfig,
+    LLMRAGQueryGeneratorConfig,
+    RAGDocument,
+    RAGQueryConfig,
+    RAGQueryGenerator,
+    RAGQueryGeneratorConfig,
+    RAGQueryResult,
+    RAGSearchMode,
+    Ranker,
+    RRFRanker,
+    WeightedRanker,
+)
+from .resource import Resource, ResourceType
+from .safety import (
+    ModerationObject,
+    ModerationObjectResults,
+    RunShieldResponse,
+    Safety,
+    SafetyViolation,
+    ShieldStore,
+    ViolationLevel,
+)
+from .schema_utils import (
+    CallableT,
+    ExtraBodyField,
+    WebMethod,
+    json_schema_type,
+    register_schema,
+    webmethod,
+)
+from .scoring import (
+    ScoreBatchResponse,
+    ScoreResponse,
+    Scoring,
+    ScoringFunctionStore,
+    ScoringResult,
+    ScoringResultRow,
+)
+from .scoring_functions import (
+    AggregationFunctionType,
+    BasicScoringFnParams,
+    CommonScoringFnFields,
+    ListScoringFunctionsResponse,
+    LLMAsJudgeScoringFnParams,
+    RegexParserScoringFnParams,
+    ScoringFn,
+    ScoringFnInput,
+    ScoringFnParams,
+    ScoringFnParamsType,
+    ScoringFunctions,
+)
+from .shields import (
+    CommonShieldFields,
+    ListShieldsResponse,
+    Shield,
+    ShieldInput,
+    Shields,
+)
+
+# Import from strong_typing
+from .strong_typing.core import JsonType
+from .strong_typing.docstring import Docstring, parse_type
+from .strong_typing.inspection import (
+    get_signature,
+    is_generic_list,
+    is_type_optional,
+    is_type_union,
+    is_unwrapped_body_param,
+    unwrap_generic_list,
+    unwrap_optional_type,
+    unwrap_union_types,
+)
+from .strong_typing.name import python_type_to_name
+from .strong_typing.schema import (
+    JsonSchemaGenerator,
+    Schema,
+    SchemaOptions,
+    StrictJsonType,
+    get_schema_identifier,
+)
+from .strong_typing.serialization import json_dump_string, object_to_json
+from .tools import (
+    ListToolDefsResponse,
+    ListToolGroupsResponse,
+    SpecialToolGroup,
+    ToolDef,
+    ToolGroup,
+    ToolGroupInput,
+    ToolGroups,
+    ToolInvocationResult,
+    ToolRuntime,
+    ToolStore,
+)
+from .vector_io import (
+    Chunk,
+    ChunkMetadata,
+    OpenAICreateVectorStoreFileBatchRequestWithExtraBody,
+    OpenAICreateVectorStoreRequestWithExtraBody,
+    QueryChunksResponse,
+    SearchRankingOptions,
+    VectorIO,
+    VectorStoreChunkingStrategy,
+    VectorStoreChunkingStrategyAuto,
+    VectorStoreChunkingStrategyStatic,
+    VectorStoreChunkingStrategyStaticConfig,
+    VectorStoreContent,
+    VectorStoreCreateRequest,
+    VectorStoreDeleteResponse,
+    VectorStoreFileBatchObject,
+    VectorStoreFileContentResponse,
+    VectorStoreFileCounts,
+    VectorStoreFileDeleteResponse,
+    VectorStoreFileLastError,
+    VectorStoreFileObject,
+    VectorStoreFilesListInBatchResponse,
+    VectorStoreFileStatus,
+    VectorStoreListFilesResponse,
+    VectorStoreListResponse,
+    VectorStoreModifyRequest,
+    VectorStoreObject,
+    VectorStoreSearchRequest,
+    VectorStoreSearchResponse,
+    VectorStoreSearchResponsePage,
+    VectorStoreTable,
+)
+from .vector_stores import VectorStore, VectorStoreInput
+from .version import (
+    LLAMA_STACK_API_V1,
+    LLAMA_STACK_API_V1ALPHA,
+    LLAMA_STACK_API_V1BETA,
+)
+
+__all__ = [
+    # Submodules
+    "common",
+    "strong_typing",
+    # Version constants
+    "LLAMA_STACK_API_V1",
+    "LLAMA_STACK_API_V1ALPHA",
+    "LLAMA_STACK_API_V1BETA",
+    # API Symbols
+    "Agents",
+    "AggregationFunctionType",
+    "AlgorithmConfig",
+    "AllowedToolsFilter",
+    "Api",
+    "ApiFilter",
+    "ApprovalFilter",
+    "BasicScoringFnParams",
+    "Batches",
+    "BatchObject",
+    "Benchmark",
+    "BenchmarkConfig",
+    "BenchmarkInput",
+    "Benchmarks",
+    "BenchmarksProtocolPrivate",
+    "Bf16QuantizationConfig",
+    "CallableT",
+    "ChatCompletionInputType",
+    "ChatCompletionResponseEventType",
+    "Checkpoint",
+    "Chunk",
+    "ChunkMetadata",
+    "CommonBenchmarkFields",
+    "ConflictError",
+    "CommonDatasetFields",
+    "CommonModelFields",
+    "CommonScoringFnFields",
+    "CommonShieldFields",
+    "CompletionInputType",
+    "CompletionRequest",
+    "Conversation",
+    "ConversationDeletedResource",
+    "ConversationItem",
+    "ConversationItemCreateRequest",
+    "ConversationItemDeletedResource",
+    "ConversationItemInclude",
+    "ConversationItemList",
+    "ConversationMessage",
+    "Conversations",
+    "DPOAlignmentConfig",
+    "DPOLossType",
+    "DataConfig",
+    "DataSource",
+    "Dataset",
+    "DatasetFormat",
+    "DatasetIO",
+    "DatasetInput",
+    "DatasetPurpose",
+    "DatasetNotFoundError",
+    "DatasetStore",
+    "DatasetType",
+    "Datasets",
+    "DatasetsProtocolPrivate",
+    "DefaultRAGQueryGeneratorConfig",
+    "Docstring",
+    "DynamicApiMeta",
+    "EfficiencyConfig",
+    "EmbeddingTaskType",
+    "EmbeddingsResponse",
+    "Error",
+    "Eval",
+    "EvalCandidate",
+    "EvaluateResponse",
+    "ExpiresAfter",
+    "ExternalApiSpec",
+    "ExtraBodyField",
+    "Files",
+    "Fp8QuantizationConfig",
+    "get_schema_identifier",
+    "get_signature",
+    "GrammarResponseFormat",
+    "GreedySamplingStrategy",
+    "HealthInfo",
+    "HealthResponse",
+    "HealthStatus",
+    "ImageContentItem",
+    "Inference",
+    "InferenceProvider",
+    "InlineProviderSpec",
+    "Inspect",
+    "Int4QuantizationConfig",
+    "InterleavedContent",
+    "InterleavedContentItem",
+    "InvalidConversationIdError",
+    "is_generic_list",
+    "is_type_optional",
+    "is_type_union",
+    "is_unwrapped_body_param",
+    "Job",
+    "JobStatus",
+    "json_dump_string",
+    "json_schema_type",
+    "JsonSchemaGenerator",
+    "JsonSchemaResponseFormat",
+    "JsonType",
+    "LLMAsJudgeScoringFnParams",
+    "LLMRAGQueryGeneratorConfig",
+    "ListBatchesResponse",
+    "ListBenchmarksResponse",
+    "ListDatasetsResponse",
+    "ListModelsResponse",
+    "ListOpenAIChatCompletionResponse",
+    "ListOpenAIFileResponse",
+    "ListOpenAIResponseInputItem",
+    "ListOpenAIResponseObject",
+    "ListPostTrainingJobsResponse",
+    "ListPromptsResponse",
+    "ListProvidersResponse",
+    "ListRoutesResponse",
+    "ListScoringFunctionsResponse",
+    "ListShieldsResponse",
+    "ListToolDefsResponse",
+    "ListToolGroupsResponse",
+    "LogProbConfig",
+    "LoraFinetuningConfig",
+    "MCPListToolsTool",
+    "Metadata",
+    "Model",
+    "ModelCandidate",
+    "ModelInput",
+    "ModelNotFoundError",
+    "ModelStore",
+    "ModelType",
+    "ModelTypeError",
+    "Models",
+    "ModelsProtocolPrivate",
+    "ModerationObject",
+    "ModerationObjectResults",
+    "NumberType",
+    "object_to_json",
+    "OpenAIAssistantMessageParam",
+    "OpenAIChatCompletion",
+    "OpenAIChatCompletionChunk",
+    "OpenAIChatCompletionContentPartImageParam",
+    "OpenAIChatCompletionContentPartParam",
+    "OpenAIChatCompletionContentPartTextParam",
+    "OpenAIChatCompletionMessageContent",
+    "OpenAIChatCompletionRequestWithExtraBody",
+    "OpenAIChatCompletionTextOnlyMessageContent",
+    "OpenAIChatCompletionToolCall",
+    "OpenAIChatCompletionToolCallFunction",
+    "OpenAIChatCompletionUsage",
+    "OpenAIChatCompletionUsageCompletionTokensDetails",
+    "OpenAIChatCompletionUsagePromptTokensDetails",
+    "OpenAIChoice",
+    "OpenAIChoiceDelta",
+    "OpenAIChoiceLogprobs",
+    "OpenAIChunkChoice",
+    "OpenAICompletion",
+    "OpenAICompletionChoice",
+    "OpenAICompletionLogprobs",
+    "OpenAICompletionRequestWithExtraBody",
+    "OpenAICompletionWithInputMessages",
+    "OpenAICreateVectorStoreFileBatchRequestWithExtraBody",
+    "OpenAICreateVectorStoreRequestWithExtraBody",
+    "OpenAIDeleteResponseObject",
+    "OpenAIDeveloperMessageParam",
+    "OpenAIEmbeddingData",
+    "OpenAIEmbeddingUsage",
+    "OpenAIEmbeddingsRequestWithExtraBody",
+    "OpenAIEmbeddingsResponse",
+    "OpenAIFile",
+    "OpenAIFileDeleteResponse",
+    "OpenAIFileFile",
+    "OpenAIFileObject",
+    "OpenAIFilePurpose",
+    "OpenAIImageURL",
+    "OpenAIJSONSchema",
+    "OpenAIListModelsResponse",
+    "OpenAIMessageParam",
+    "OpenAIModel",
+    "Order",
+    "OpenAIResponseAnnotationCitation",
+    "OpenAIResponseAnnotationContainerFileCitation",
+    "OpenAIResponseAnnotationFileCitation",
+    "OpenAIResponseAnnotationFilePath",
+    "OpenAIResponseAnnotations",
+    "OpenAIResponseContentPart",
+    "OpenAIResponseContentPartOutputText",
+    "OpenAIResponseContentPartReasoningSummary",
+    "OpenAIResponseContentPartReasoningText",
+    "OpenAIResponseContentPartRefusal",
+    "OpenAIResponseError",
+    "OpenAIResponseFormatJSONObject",
+    "OpenAIResponseFormatJSONSchema",
+    "OpenAIResponseFormatParam",
+    "OpenAIResponseFormatText",
+    "OpenAIResponseInput",
+    "OpenAIResponseInputFunctionToolCallOutput",
+    "OpenAIResponseInputMessageContent",
+    "OpenAIResponseInputMessageContentFile",
+    "OpenAIResponseInputMessageContentImage",
+    "OpenAIResponseInputMessageContentText",
+    "OpenAIResponseInputTool",
+    "OpenAIResponseInputToolFileSearch",
+    "OpenAIResponseInputToolFunction",
+    "OpenAIResponseInputToolMCP",
+    "OpenAIResponseInputToolWebSearch",
+    "OpenAIResponseMCPApprovalRequest",
+    "OpenAIResponseMCPApprovalResponse",
+    "OpenAIResponseMessage",
+    "OpenAIResponseObject",
+    "OpenAIResponseObjectStream",
+    "OpenAIResponseObjectStreamResponseCompleted",
+    "OpenAIResponseObjectStreamResponseContentPartAdded",
+    "OpenAIResponseObjectStreamResponseContentPartDone",
+    "OpenAIResponseObjectStreamResponseCreated",
+    "OpenAIResponseObjectStreamResponseFailed",
+    "OpenAIResponseObjectStreamResponseFileSearchCallCompleted",
+    "OpenAIResponseObjectStreamResponseFileSearchCallInProgress",
+    "OpenAIResponseObjectStreamResponseFileSearchCallSearching",
+    "OpenAIResponseObjectStreamResponseFunctionCallArgumentsDelta",
+    "OpenAIResponseObjectStreamResponseFunctionCallArgumentsDone",
+    "OpenAIResponseObjectStreamResponseInProgress",
+    "OpenAIResponseObjectStreamResponseIncomplete",
+    "OpenAIResponseObjectStreamResponseMcpCallArgumentsDelta",
+    "OpenAIResponseObjectStreamResponseMcpCallArgumentsDone",
+    "OpenAIResponseObjectStreamResponseMcpCallCompleted",
+    "OpenAIResponseObjectStreamResponseMcpCallFailed",
+    "OpenAIResponseObjectStreamResponseMcpCallInProgress",
+    "OpenAIResponseObjectStreamResponseMcpListToolsCompleted",
+    "OpenAIResponseObjectStreamResponseMcpListToolsFailed",
+    "OpenAIResponseObjectStreamResponseMcpListToolsInProgress",
+    "OpenAIResponseObjectStreamResponseOutputItemAdded",
+    "OpenAIResponseObjectStreamResponseOutputItemDone",
+    "OpenAIResponseObjectStreamResponseOutputTextAnnotationAdded",
+    "OpenAIResponseObjectStreamResponseOutputTextDelta",
+    "OpenAIResponseObjectStreamResponseOutputTextDone",
+    "OpenAIResponseObjectStreamResponseReasoningSummaryPartAdded",
+    "OpenAIResponseObjectStreamResponseReasoningSummaryPartDone",
+    "OpenAIResponseObjectStreamResponseReasoningSummaryTextDelta",
+    "OpenAIResponseObjectStreamResponseReasoningSummaryTextDone",
+    "OpenAIResponseObjectStreamResponseReasoningTextDelta",
+    "OpenAIResponseObjectStreamResponseReasoningTextDone",
+    "OpenAIResponseObjectStreamResponseRefusalDelta",
+    "OpenAIResponseObjectStreamResponseRefusalDone",
+    "OpenAIResponseObjectStreamResponseWebSearchCallCompleted",
+    "OpenAIResponseObjectStreamResponseWebSearchCallInProgress",
+    "OpenAIResponseObjectStreamResponseWebSearchCallSearching",
+    "OpenAIResponseObjectWithInput",
+    "OpenAIResponseOutput",
+    "OpenAIResponseOutputMessageContent",
+    "OpenAIResponseOutputMessageContentOutputText",
+    "OpenAIResponseOutputMessageFileSearchToolCall",
+    "OpenAIResponseOutputMessageFileSearchToolCallResults",
+    "OpenAIResponseOutputMessageFunctionToolCall",
+    "OpenAIResponseOutputMessageMCPCall",
+    "OpenAIResponseOutputMessageMCPListTools",
+    "OpenAIResponseOutputMessageWebSearchToolCall",
+    "OpenAIResponsePrompt",
+    "OpenAIResponseText",
+    "OpenAIResponseTextFormat",
+    "OpenAIResponseTool",
+    "OpenAIResponseToolMCP",
+    "OpenAIResponseUsage",
+    "OpenAIResponseUsageInputTokensDetails",
+    "OpenAIResponseUsageOutputTokensDetails",
+    "OpenAISystemMessageParam",
+    "OpenAITokenLogProb",
+    "OpenAIToolMessageParam",
+    "OpenAITopLogProb",
+    "OpenAIUserMessageParam",
+    "OptimizerConfig",
+    "OptimizerType",
+    "PaginatedResponse",
+    "ParamType",
+    "parse_type",
+    "PostTraining",
+    "PostTrainingMetric",
+    "PostTrainingJob",
+    "PostTrainingJobArtifactsResponse",
+    "PostTrainingJobLogStream",
+    "PostTrainingJobStatusResponse",
+    "PostTrainingRLHFRequest",
+    "Prompt",
+    "Prompts",
+    "ProviderInfo",
+    "ProviderSpec",
+    "Providers",
+    "python_type_to_name",
+    "QATFinetuningConfig",
+    "QuantizationConfig",
+    "QuantizationType",
+    "QueryChunksResponse",
+    "RAGDocument",
+    "RAGQueryConfig",
+    "RAGQueryGenerator",
+    "RAGQueryGeneratorConfig",
+    "RAGQueryResult",
+    "RAGSearchMode",
+    "register_schema",
+    "RLHFAlgorithm",
+    "RRFRanker",
+    "Ranker",
+    "RegexParserScoringFnParams",
+    "RemoteProviderConfig",
+    "RemoteProviderSpec",
+    "RerankData",
+    "RerankResponse",
+    "Resource",
+    "ResourceNotFoundError",
+    "ResourceType",
+    "ResponseFormat",
+    "ResponseFormatType",
+    "ResponseGuardrail",
+    "ResponseGuardrailSpec",
+    "RouteInfo",
+    "RoutingTable",
+    "RowsDataSource",
+    "RunShieldResponse",
+    "Safety",
+    "SafetyViolation",
+    "SamplingParams",
+    "SamplingStrategy",
+    "ScoreBatchResponse",
+    "ScoreResponse",
+    "Scoring",
+    "ScoringFn",
+    "ScoringFnInput",
+    "ScoringFnParams",
+    "ScoringFnParamsType",
+    "ScoringFunctionStore",
+    "ScoringFunctions",
+    "ScoringFunctionsProtocolPrivate",
+    "ScoringResult",
+    "ScoringResultRow",
+    "Schema",
+    "SchemaOptions",
+    "SearchRankingOptions",
+    "Shield",
+    "ShieldInput",
+    "ShieldStore",
+    "Shields",
+    "ShieldsProtocolPrivate",
+    "SpecialToolGroup",
+    "StrictJsonType",
+    "StringType",
+    "SystemMessage",
+    "SystemMessageBehavior",
+    "TextContentItem",
+    "TextTruncation",
+    "TokenLogProbs",
+    "TokenValidationError",
+    "ToolChoice",
+    "ToolGroupNotFoundError",
+    "ToolDef",
+    "ToolGroup",
+    "ToolGroupInput",
+    "ToolGroups",
+    "ToolGroupsProtocolPrivate",
+    "ToolInvocationResult",
+    "ToolResponseMessage",
+    "ToolRuntime",
+    "ToolStore",
+    "TopKSamplingStrategy",
+    "TopPSamplingStrategy",
+    "TrainingConfig",
+    "UnsupportedModelError",
+    "unwrap_generic_list",
+    "unwrap_optional_type",
+    "unwrap_union_types",
+    "URIDataSource",
+    "URL",
+    "_URLOrData",
+    "UserMessage",
+    "VectorIO",
+    "VectorStore",
+    "VectorStoreChunkingStrategy",
+    "VectorStoreChunkingStrategyAuto",
+    "VectorStoreChunkingStrategyStatic",
+    "VectorStoreChunkingStrategyStaticConfig",
+    "VectorStoreContent",
+    "VectorStoreCreateRequest",
+    "VectorStoreDeleteResponse",
+    "VectorStoreFileBatchObject",
+    "VectorStoreFileContentResponse",
+    "VectorStoreFileCounts",
+    "VectorStoreFileDeleteResponse",
+    "VectorStoreFileLastError",
+    "VectorStoreFileObject",
+    "VectorStoreFileStatus",
+    "VectorStoreFilesListInBatchResponse",
+    "VectorStoreInput",
+    "VectorStoreListFilesResponse",
+    "VectorStoreListResponse",
+    "VectorStoreModifyRequest",
+    "VectorStoreObject",
+    "VectorStoreSearchRequest",
+    "VectorStoreSearchResponse",
+    "VectorStoreSearchResponsePage",
+    "VectorStoreTable",
+    "VectorStoreNotFoundError",
+    "VectorStoresProtocolPrivate",
+    "VersionInfo",
+    "ViolationLevel",
+    "webmethod",
+    "WebMethod",
+    "WebSearchToolTypes",
+    "WeightedRanker",
+]

src/llama_stack/cli/stack/_list_deps.py

@@ -9,7 +9,7 @@ import sys
 from pathlib import Path
 
 import yaml
-from llama_stack_api.datatypes import Api
+from llama_stack_api import Api
 from termcolor import cprint
 
 from llama_stack.cli.stack.utils import ImageType

src/llama_stack/cli/stack/utils.py

@@ -11,7 +11,7 @@ from functools import lru_cache
 from pathlib import Path
 
 import yaml
-from llama_stack_api.datatypes import Api
+from llama_stack_api import Api
 from termcolor import cprint
 
 from llama_stack.core.datatypes import (

src/llama_stack/core/build.py

@@ -6,7 +6,7 @@
 
 import sys
 
-from llama_stack_api.datatypes import Api
+from llama_stack_api import Api
 from pydantic import BaseModel
 from termcolor import cprint
 

src/llama_stack/core/client.py

@@ -12,7 +12,7 @@ from enum import Enum
 from typing import Any, Union, get_args, get_origin
 
 import httpx
-from llama_stack_api.datatypes import RemoteProviderConfig
+from llama_stack_api import RemoteProviderConfig
 from pydantic import BaseModel, parse_obj_as
 from termcolor import cprint
 

src/llama_stack/core/configure.py

@@ -6,7 +6,7 @@
 import textwrap
 from typing import Any
 
-from llama_stack_api.datatypes import Api, ProviderSpec
+from llama_stack_api import Api, ProviderSpec
 
 from llama_stack.core.datatypes import (
     LLAMA_STACK_RUN_CONFIG_VERSION,

src/llama_stack/core/conversations/conversations.py

@@ -8,7 +8,7 @@ import secrets
 import time
 from typing import Any, Literal
 
-from llama_stack_api.conversations import (
+from llama_stack_api import (
     Conversation,
     ConversationDeletedResource,
     ConversationItem,

src/llama_stack/core/datatypes.py

@@ -9,21 +9,32 @@ from pathlib import Path
 from typing import Annotated, Any, Literal, Self
 from urllib.parse import urlparse
 
-from llama_stack_api.benchmarks import Benchmark, BenchmarkInput
+from llama_stack_api import (
-from llama_stack_api.datasetio import DatasetIO
+    Api,
-from llama_stack_api.datasets import Dataset, DatasetInput
+    Benchmark,
-from llama_stack_api.datatypes import Api, ProviderSpec
+    BenchmarkInput,
-from llama_stack_api.eval import Eval
+    Dataset,
-from llama_stack_api.inference import Inference
+    DatasetInput,
-from llama_stack_api.models import Model, ModelInput
+    DatasetIO,
-from llama_stack_api.resource import Resource
+    Eval,
-from llama_stack_api.safety import Safety
+    Inference,
-from llama_stack_api.scoring import Scoring
+    Model,
-from llama_stack_api.scoring_functions import ScoringFn, ScoringFnInput
+    ModelInput,
-from llama_stack_api.shields import Shield, ShieldInput
+    ProviderSpec,
-from llama_stack_api.tools import ToolGroup, ToolGroupInput, ToolRuntime
+    Resource,
-from llama_stack_api.vector_io import VectorIO
+    Safety,
-from llama_stack_api.vector_stores import VectorStore, VectorStoreInput
+    Scoring,
+    ScoringFn,
+    ScoringFnInput,
+    Shield,
+    ShieldInput,
+    ToolGroup,
+    ToolGroupInput,
+    ToolRuntime,
+    VectorIO,
+    VectorStore,
+    VectorStoreInput,
+)
 from pydantic import BaseModel, Field, field_validator, model_validator
 
 from llama_stack.core.access_control.datatypes import AccessRule

src/llama_stack/core/distribution.py

@@ -10,7 +10,7 @@ import os
 from typing import Any
 
 import yaml
-from llama_stack_api.datatypes import (
+from llama_stack_api import (
     Api,
     InlineProviderSpec,
     ProviderSpec,

src/llama_stack/core/external.py

@@ -6,7 +6,7 @@
 
 
 import yaml
-from llama_stack_api.datatypes import Api, ExternalApiSpec
+from llama_stack_api import Api, ExternalApiSpec
 
 from llama_stack.core.datatypes import BuildConfig, StackRunConfig
 from llama_stack.log import get_logger

src/llama_stack/core/inspect.py

@@ -6,9 +6,9 @@
 
 from importlib.metadata import version
 
-from llama_stack_api.datatypes import HealthStatus
+from llama_stack_api import (
-from llama_stack_api.inspect import (
     HealthInfo,
+    HealthStatus,
     Inspect,
     ListRoutesResponse,
     RouteInfo,

src/llama_stack/core/library_client.py

@@ -18,7 +18,7 @@ from typing import Any, TypeVar, Union, get_args, get_origin
 import httpx
 import yaml
 from fastapi import Response as FastAPIResponse
-from llama_stack_api.strong_typing.inspection import is_unwrapped_body_param
+from llama_stack_api import is_unwrapped_body_param
 
 try:
     from llama_stack_client import (

src/llama_stack/core/prompts/prompts.py

@@ -7,7 +7,7 @@
 import json
 from typing import Any
 
-from llama_stack_api.prompts import ListPromptsResponse, Prompt, Prompts
+from llama_stack_api import ListPromptsResponse, Prompt, Prompts
 from pydantic import BaseModel
 
 from llama_stack.core.datatypes import StackRunConfig

src/llama_stack/core/providers.py

@@ -7,8 +7,7 @@
 import asyncio
 from typing import Any
 
-from llama_stack_api.datatypes import HealthResponse, HealthStatus
+from llama_stack_api import HealthResponse, HealthStatus, ListProvidersResponse, ProviderInfo, Providers
-from llama_stack_api.providers import ListProvidersResponse, ProviderInfo, Providers
 from pydantic import BaseModel
 
 from llama_stack.log import get_logger

src/llama_stack/core/resolver.py

@@ -8,41 +8,45 @@ import importlib.metadata
 import inspect
 from typing import Any
 
-from llama_stack_api.agents import Agents
+from llama_stack_api import (
-from llama_stack_api.batches import Batches
+    LLAMA_STACK_API_V1ALPHA,
-from llama_stack_api.benchmarks import Benchmarks
+    Agents,
-from llama_stack_api.conversations import Conversations
-from llama_stack_api.datasetio import DatasetIO
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.datatypes import (
     Api,
+    Batches,
+    Benchmarks,
     BenchmarksProtocolPrivate,
+    Conversations,
+    DatasetIO,
+    Datasets,
     DatasetsProtocolPrivate,
+    Eval,
     ExternalApiSpec,
+    Files,
+    Inference,
+    InferenceProvider,
+    Inspect,
+    Models,
     ModelsProtocolPrivate,
+    PostTraining,
+    Prompts,
     ProviderSpec,
     RemoteProviderConfig,
     RemoteProviderSpec,
+    Safety,
+    Scoring,
+    ScoringFunctions,
     ScoringFunctionsProtocolPrivate,
+    Shields,
     ShieldsProtocolPrivate,
+    ToolGroups,
     ToolGroupsProtocolPrivate,
+    ToolRuntime,
+    VectorIO,
+    VectorStore,
+)
+from llama_stack_api import (
+    Providers as ProvidersAPI,
 )
-from llama_stack_api.eval import Eval
-from llama_stack_api.files import Files
-from llama_stack_api.inference import Inference, InferenceProvider
-from llama_stack_api.inspect import Inspect
-from llama_stack_api.models import Models
-from llama_stack_api.post_training import PostTraining
-from llama_stack_api.prompts import Prompts
-from llama_stack_api.providers import Providers as ProvidersAPI
-from llama_stack_api.safety import Safety
-from llama_stack_api.scoring import Scoring
-from llama_stack_api.scoring_functions import ScoringFunctions
-from llama_stack_api.shields import Shields
-from llama_stack_api.tools import ToolGroups, ToolRuntime
-from llama_stack_api.vector_io import VectorIO
-from llama_stack_api.vector_stores import VectorStore
-from llama_stack_api.version import LLAMA_STACK_API_V1ALPHA
 
 from llama_stack.core.client import get_client_impl
 from llama_stack.core.datatypes import (

src/llama_stack/core/routers/__init__.py

@@ -6,7 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.datatypes import Api, RoutingTable
+from llama_stack_api import Api, RoutingTable
 
 from llama_stack.core.datatypes import (
     AccessRule,

src/llama_stack/core/routers/datasets.py

@@ -6,10 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.common.responses import PaginatedResponse
+from llama_stack_api import DatasetIO, DatasetPurpose, DataSource, PaginatedResponse, RoutingTable
-from llama_stack_api.datasetio import DatasetIO
-from llama_stack_api.datasets import DatasetPurpose, DataSource
-from llama_stack_api.datatypes import RoutingTable
 
 from llama_stack.log import get_logger
 

src/llama_stack/core/routers/eval_scoring.py

@@ -6,9 +6,12 @@
 
 from typing import Any
 
-from llama_stack_api.datatypes import RoutingTable
+from llama_stack_api import (
-from llama_stack_api.eval import BenchmarkConfig, Eval, EvaluateResponse, Job
+    BenchmarkConfig,
-from llama_stack_api.scoring import (
+    Eval,
+    EvaluateResponse,
+    Job,
+    RoutingTable,
     ScoreBatchResponse,
     ScoreResponse,
     Scoring,

src/llama_stack/core/routers/inference.py

@@ -11,11 +11,14 @@ from datetime import UTC, datetime
 from typing import Annotated, Any
 
 from fastapi import Body
-from llama_stack_api.common.errors import ModelNotFoundError, ModelTypeError
+from llama_stack_api import (
-from llama_stack_api.datatypes import HealthResponse, HealthStatus, RoutingTable
+    HealthResponse,
-from llama_stack_api.inference import (
+    HealthStatus,
     Inference,
     ListOpenAIChatCompletionResponse,
+    ModelNotFoundError,
+    ModelType,
+    ModelTypeError,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionChunk,
@@ -34,8 +37,8 @@ from llama_stack_api.inference import (
     OpenAIMessageParam,
     Order,
     RerankResponse,
+    RoutingTable,
 )
-from llama_stack_api.models import ModelType
 from openai.types.chat import ChatCompletionToolChoiceOptionParam as OpenAIChatCompletionToolChoiceOptionParam
 from openai.types.chat import ChatCompletionToolParam as OpenAIChatCompletionToolParam
 from pydantic import TypeAdapter

src/llama_stack/core/routers/safety.py

@@ -6,10 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.datatypes import RoutingTable
+from llama_stack_api import ModerationObject, OpenAIMessageParam, RoutingTable, RunShieldResponse, Safety, Shield
-from llama_stack_api.inference import OpenAIMessageParam
-from llama_stack_api.safety import ModerationObject, RunShieldResponse, Safety
-from llama_stack_api.shields import Shield
 
 from llama_stack.core.datatypes import SafetyConfig
 from llama_stack.log import get_logger

src/llama_stack/core/routers/tool_runtime.py

@@ -6,10 +6,8 @@
 
 from typing import Any
 
-from llama_stack_api.common.content_types import (
+from llama_stack_api import (
     URL,
-)
-from llama_stack_api.tools import (
     ListToolDefsResponse,
     ToolRuntime,
 )

src/llama_stack/core/routers/vector_io.py

@@ -9,14 +9,16 @@ import uuid
 from typing import Annotated, Any
 
 from fastapi import Body
-from llama_stack_api.common.content_types import InterleavedContent
+from llama_stack_api import (
-from llama_stack_api.datatypes import HealthResponse, HealthStatus, RoutingTable
-from llama_stack_api.models import ModelType
-from llama_stack_api.vector_io import (
     Chunk,
+    HealthResponse,
+    HealthStatus,
+    InterleavedContent,
+    ModelType,
     OpenAICreateVectorStoreFileBatchRequestWithExtraBody,
     OpenAICreateVectorStoreRequestWithExtraBody,
     QueryChunksResponse,
+    RoutingTable,
     SearchRankingOptions,
     VectorIO,
     VectorStoreChunkingStrategy,

src/llama_stack/core/routing_tables/benchmarks.py

@@ -6,7 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.benchmarks import Benchmark, Benchmarks, ListBenchmarksResponse
+from llama_stack_api import Benchmark, Benchmarks, ListBenchmarksResponse
 
 from llama_stack.core.datatypes import (
     BenchmarkWithOwner,

src/llama_stack/core/routing_tables/common.py

@@ -6,10 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.common.errors import ModelNotFoundError
+from llama_stack_api import Api, Model, ModelNotFoundError, ResourceType, RoutingTable
-from llama_stack_api.datatypes import Api, RoutingTable
-from llama_stack_api.models import Model
-from llama_stack_api.resource import ResourceType
 
 from llama_stack.core.access_control.access_control import AccessDeniedError, is_action_allowed
 from llama_stack.core.access_control.datatypes import Action

src/llama_stack/core/routing_tables/datasets.py

@@ -7,18 +7,18 @@
 import uuid
 from typing import Any
 
-from llama_stack_api.common.errors import DatasetNotFoundError
+from llama_stack_api import (
-from llama_stack_api.datasets import (
     Dataset,
+    DatasetNotFoundError,
     DatasetPurpose,
     Datasets,
     DatasetType,
     DataSource,
     ListDatasetsResponse,
+    ResourceType,
     RowsDataSource,
     URIDataSource,
 )
-from llama_stack_api.resource import ResourceType
 
 from llama_stack.core.datatypes import (
     DatasetWithOwner,

src/llama_stack/core/routing_tables/models.py

@@ -7,10 +7,10 @@
 import time
 from typing import Any
 
-from llama_stack_api.common.errors import ModelNotFoundError
+from llama_stack_api import (
-from llama_stack_api.models import (
     ListModelsResponse,
     Model,
+    ModelNotFoundError,
     Models,
     ModelType,
     OpenAIListModelsResponse,

src/llama_stack/core/routing_tables/scoring_functions.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import ParamType
+from llama_stack_api import (
-from llama_stack_api.resource import ResourceType
-from llama_stack_api.scoring_functions import (
     ListScoringFunctionsResponse,
+    ParamType,
+    ResourceType,
     ScoringFn,
     ScoringFnParams,
     ScoringFunctions,

src/llama_stack/core/routing_tables/shields.py

@@ -6,8 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.resource import ResourceType
+from llama_stack_api import ListShieldsResponse, ResourceType, Shield, Shields
-from llama_stack_api.shields import ListShieldsResponse, Shield, Shields
 
 from llama_stack.core.datatypes import (
     ShieldWithOwner,

src/llama_stack/core/routing_tables/toolgroups.py

@@ -6,9 +6,15 @@
 
 from typing import Any
 
-from llama_stack_api.common.content_types import URL
+from llama_stack_api import (
-from llama_stack_api.common.errors import ToolGroupNotFoundError
+    URL,
-from llama_stack_api.tools import ListToolDefsResponse, ListToolGroupsResponse, ToolDef, ToolGroup, ToolGroups
+    ListToolDefsResponse,
+    ListToolGroupsResponse,
+    ToolDef,
+    ToolGroup,
+    ToolGroupNotFoundError,
+    ToolGroups,
+)
 
 from llama_stack.core.datatypes import AuthenticationRequiredError, ToolGroupWithOwner
 from llama_stack.log import get_logger

src/llama_stack/core/routing_tables/vector_stores.py

@@ -6,12 +6,12 @@
 
 from typing import Any
 
-from llama_stack_api.common.errors import ModelNotFoundError, ModelTypeError
-from llama_stack_api.models import ModelType
-from llama_stack_api.resource import ResourceType
-
 # Removed VectorStores import to avoid exposing public API
-from llama_stack_api.vector_io import (
+from llama_stack_api import (
+    ModelNotFoundError,
+    ModelType,
+    ModelTypeError,
+    ResourceType,
     SearchRankingOptions,
     VectorStoreChunkingStrategy,
     VectorStoreDeleteResponse,

src/llama_stack/core/server/auth_providers.py

@@ -11,7 +11,7 @@ from urllib.parse import parse_qs, urljoin, urlparse
 
 import httpx
 import jwt
-from llama_stack_api.common.errors import TokenValidationError
+from llama_stack_api import TokenValidationError
 from pydantic import BaseModel, Field
 
 from llama_stack.core.datatypes import (

src/llama_stack/core/server/routes.py

@@ -10,8 +10,7 @@ from collections.abc import Callable
 from typing import Any
 
 from aiohttp import hdrs
-from llama_stack_api.datatypes import Api, ExternalApiSpec
+from llama_stack_api import Api, ExternalApiSpec, WebMethod
-from llama_stack_api.schema_utils import WebMethod
 from starlette.routing import Route
 
 from llama_stack.core.resolver import api_protocol_map

src/llama_stack/core/server/server.py

@@ -28,9 +28,7 @@ from fastapi import Path as FastapiPath
 from fastapi.exceptions import RequestValidationError
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse, StreamingResponse
-from llama_stack_api.common.errors import ConflictError, ResourceNotFoundError
+from llama_stack_api import Api, ConflictError, PaginatedResponse, ResourceNotFoundError
-from llama_stack_api.common.responses import PaginatedResponse
-from llama_stack_api.datatypes import Api
 from openai import BadRequestError
 from pydantic import BaseModel, ValidationError
 

src/llama_stack/core/stack.py

@@ -12,27 +12,30 @@ import tempfile
 from typing import Any
 
 import yaml
-from llama_stack_api.agents import Agents
+from llama_stack_api import (
-from llama_stack_api.batches import Batches
+    Agents,
-from llama_stack_api.benchmarks import Benchmarks
+    Api,
-from llama_stack_api.conversations import Conversations
+    Batches,
-from llama_stack_api.datasetio import DatasetIO
+    Benchmarks,
-from llama_stack_api.datasets import Datasets
+    Conversations,
-from llama_stack_api.datatypes import Api
+    DatasetIO,
-from llama_stack_api.eval import Eval
+    Datasets,
-from llama_stack_api.files import Files
+    Eval,
-from llama_stack_api.inference import Inference
+    Files,
-from llama_stack_api.inspect import Inspect
+    Inference,
-from llama_stack_api.models import Models
+    Inspect,
-from llama_stack_api.post_training import PostTraining
+    Models,
-from llama_stack_api.prompts import Prompts
+    PostTraining,
-from llama_stack_api.providers import Providers
+    Prompts,
-from llama_stack_api.safety import Safety
+    Providers,
-from llama_stack_api.scoring import Scoring
+    Safety,
-from llama_stack_api.scoring_functions import ScoringFunctions
+    Scoring,
-from llama_stack_api.shields import Shields
+    ScoringFunctions,
-from llama_stack_api.tools import ToolGroups, ToolRuntime
+    Shields,
-from llama_stack_api.vector_io import VectorIO
+    ToolGroups,
+    ToolRuntime,
+    VectorIO,
+)
 
 from llama_stack.core.conversations.conversations import ConversationServiceConfig, ConversationServiceImpl
 from llama_stack.core.datatypes import Provider, SafetyConfig, StackRunConfig, VectorStoresConfig

src/llama_stack/core/telemetry/telemetry.py

@@ -16,7 +16,7 @@ from typing import (
     cast,
 )
 
-from llama_stack_api.schema_utils import json_schema_type, register_schema
+from llama_stack_api import json_schema_type, register_schema
 from opentelemetry import metrics, trace
 from opentelemetry.exporter.otlp.proto.http.metric_exporter import OTLPMetricExporter
 from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter

src/llama_stack/distributions/dell/dell.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.models import ModelType
+from llama_stack_api import ModelType
 
 from llama_stack.core.datatypes import (
     BuildProvider,

src/llama_stack/distributions/meta-reference-gpu/meta_reference.py

@@ -6,7 +6,7 @@
 
 from pathlib import Path
 
-from llama_stack_api.models import ModelType
+from llama_stack_api import ModelType
 
 from llama_stack.core.datatypes import (
     BuildProvider,

src/llama_stack/distributions/open-benchmark/open_benchmark.py

@@ -5,8 +5,7 @@
 # the root directory of this source tree.
 
 
-from llama_stack_api.datasets import DatasetPurpose, URIDataSource
+from llama_stack_api import DatasetPurpose, ModelType, URIDataSource
-from llama_stack_api.models import ModelType
 
 from llama_stack.core.datatypes import (
     BenchmarkInput,

src/llama_stack/distributions/starter/starter.py

@@ -7,7 +7,7 @@
 
 from typing import Any
 
-from llama_stack_api.datatypes import RemoteProviderSpec
+from llama_stack_api import RemoteProviderSpec
 
 from llama_stack.core.datatypes import (
     BuildProvider,

src/llama_stack/distributions/template.py

@@ -10,8 +10,7 @@ from typing import Any, Literal
 import jinja2
 import rich
 import yaml
-from llama_stack_api.datasets import DatasetPurpose
+from llama_stack_api import DatasetPurpose, ModelType
-from llama_stack_api.models import ModelType
 from pydantic import BaseModel, Field
 
 from llama_stack.core.datatypes import (

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

@@ -5,25 +5,25 @@
 # the root directory of this source tree.
 
 
-from llama_stack_api.agents import (
+from llama_stack_api import (
     Agents,
+    Conversations,
+    Inference,
     ListOpenAIResponseInputItem,
     ListOpenAIResponseObject,
     OpenAIDeleteResponseObject,
     OpenAIResponseInput,
     OpenAIResponseInputTool,
     OpenAIResponseObject,
+    OpenAIResponsePrompt,
+    OpenAIResponseText,
     Order,
     ResponseGuardrail,
+    Safety,
+    ToolGroups,
+    ToolRuntime,
+    VectorIO,
 )
-from llama_stack_api.conversations import Conversations
-from llama_stack_api.inference import (
-    Inference,
-)
-from llama_stack_api.openai_responses import OpenAIResponsePrompt, OpenAIResponseText
-from llama_stack_api.safety import Safety
-from llama_stack_api.tools import ToolGroups, ToolRuntime
-from llama_stack_api.vector_io import VectorIO
 
 from llama_stack.core.datatypes import AccessRule
 from llama_stack.log import get_logger

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

@@ -8,20 +8,15 @@ import time
 import uuid
 from collections.abc import AsyncIterator
 
-from llama_stack_api.agents import Order, ResponseGuardrailSpec
+from llama_stack_api import (
-from llama_stack_api.common.errors import (
+    ConversationItem,
-    InvalidConversationIdError,
+    Conversations,
-)
-from llama_stack_api.conversations import ConversationItem, Conversations
-from llama_stack_api.inference import (
     Inference,
-    OpenAIMessageParam,
+    InvalidConversationIdError,
-    OpenAISystemMessageParam,
-)
-from llama_stack_api.openai_responses import (
     ListOpenAIResponseInputItem,
     ListOpenAIResponseObject,
     OpenAIDeleteResponseObject,
+    OpenAIMessageParam,
     OpenAIResponseInput,
     OpenAIResponseInputMessageContentText,
     OpenAIResponseInputTool,
@@ -31,10 +26,14 @@ from llama_stack_api.openai_responses import (
     OpenAIResponsePrompt,
     OpenAIResponseText,
     OpenAIResponseTextFormat,
+    OpenAISystemMessageParam,
+    Order,
+    ResponseGuardrailSpec,
+    Safety,
+    ToolGroups,
+    ToolRuntime,
+    VectorIO,
 )
-from llama_stack_api.safety import Safety
-from llama_stack_api.tools import ToolGroups, ToolRuntime
-from llama_stack_api.vector_io import VectorIO
 from pydantic import BaseModel, TypeAdapter
 
 from llama_stack.log import get_logger

src/llama_stack/providers/inline/agents/meta_reference/responses/streaming.py

@@ -8,8 +8,11 @@ import uuid
 from collections.abc import AsyncIterator
 from typing import Any
 
-from llama_stack_api.inference import (
+from llama_stack_api import (
+    AllowedToolsFilter,
+    ApprovalFilter,
     Inference,
+    MCPListToolsTool,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionChunk,
@@ -17,11 +20,6 @@ from llama_stack_api.inference import (
     OpenAIChatCompletionToolCall,
     OpenAIChoice,
     OpenAIMessageParam,
-)
-from llama_stack_api.openai_responses import (
-    AllowedToolsFilter,
-    ApprovalFilter,
-    MCPListToolsTool,
     OpenAIResponseContentPartOutputText,
     OpenAIResponseContentPartReasoningText,
     OpenAIResponseContentPartRefusal,
@@ -1024,7 +1022,7 @@ class StreamingResponseOrchestrator:
         self, tools: list[OpenAIResponseInputTool], output_messages: list[OpenAIResponseOutput]
     ) -> AsyncIterator[OpenAIResponseObjectStream]:
         """Process all tools and emit appropriate streaming events."""
-        from llama_stack_api.tools import ToolDef
+        from llama_stack_api import ToolDef
         from openai.types.chat import ChatCompletionToolParam
 
         from llama_stack.models.llama.datatypes import ToolDefinition

src/llama_stack/providers/inline/agents/meta_reference/responses/tool_executor.py

@@ -9,18 +9,12 @@ import json
 from collections.abc import AsyncIterator
 from typing import Any
 
-from llama_stack_api.common.content_types import (
+from llama_stack_api import (
     ImageContentItem,
-    TextContentItem,
-)
-from llama_stack_api.inference import (
     OpenAIChatCompletionContentPartImageParam,
     OpenAIChatCompletionContentPartTextParam,
     OpenAIChatCompletionToolCall,
     OpenAIImageURL,
-    OpenAIToolMessageParam,
-)
-from llama_stack_api.openai_responses import (
     OpenAIResponseInputToolFileSearch,
     OpenAIResponseInputToolMCP,
     OpenAIResponseObjectStreamResponseFileSearchCallCompleted,
@@ -35,9 +29,13 @@ from llama_stack_api.openai_responses import (
     OpenAIResponseOutputMessageFileSearchToolCall,
     OpenAIResponseOutputMessageFileSearchToolCallResults,
     OpenAIResponseOutputMessageWebSearchToolCall,
+    OpenAIToolMessageParam,
+    TextContentItem,
+    ToolGroups,
+    ToolInvocationResult,
+    ToolRuntime,
+    VectorIO,
 )
-from llama_stack_api.tools import ToolGroups, ToolInvocationResult, ToolRuntime
-from llama_stack_api.vector_io import VectorIO
 
 from llama_stack.core.telemetry import tracing
 from llama_stack.log import get_logger
@@ -398,7 +396,7 @@ class ToolExecutor:
         # Build output message
         message: Any
         if mcp_tool_to_server and function.name in mcp_tool_to_server:
-            from llama_stack_api.openai_responses import (
+            from llama_stack_api import (
                 OpenAIResponseOutputMessageMCPCall,
             )
 

src/llama_stack/providers/inline/agents/meta_reference/responses/types.py

@@ -7,8 +7,10 @@
 from dataclasses import dataclass
 from typing import cast
 
-from llama_stack_api.inference import OpenAIChatCompletionToolCall, OpenAIMessageParam, OpenAIResponseFormatParam
+from llama_stack_api import (
-from llama_stack_api.openai_responses import (
+    OpenAIChatCompletionToolCall,
+    OpenAIMessageParam,
+    OpenAIResponseFormatParam,
     OpenAIResponseInput,
     OpenAIResponseInputTool,
     OpenAIResponseInputToolFileSearch,

src/llama_stack/providers/inline/agents/meta_reference/responses/utils.py

@@ -9,8 +9,7 @@ import re
 import uuid
 from collections.abc import Sequence
 
-from llama_stack_api.agents import ResponseGuardrailSpec
+from llama_stack_api import (
-from llama_stack_api.inference import (
     OpenAIAssistantMessageParam,
     OpenAIChatCompletionContentPartImageParam,
     OpenAIChatCompletionContentPartParam,
@@ -22,16 +21,11 @@ from llama_stack_api.inference import (
     OpenAIImageURL,
     OpenAIJSONSchema,
     OpenAIMessageParam,
+    OpenAIResponseAnnotationFileCitation,
     OpenAIResponseFormatJSONObject,
     OpenAIResponseFormatJSONSchema,
     OpenAIResponseFormatParam,
     OpenAIResponseFormatText,
-    OpenAISystemMessageParam,
-    OpenAIToolMessageParam,
-    OpenAIUserMessageParam,
-)
-from llama_stack_api.openai_responses import (
-    OpenAIResponseAnnotationFileCitation,
     OpenAIResponseInput,
     OpenAIResponseInputFunctionToolCallOutput,
     OpenAIResponseInputMessageContent,
@@ -47,8 +41,12 @@ from llama_stack_api.openai_responses import (
     OpenAIResponseOutputMessageMCPCall,
     OpenAIResponseOutputMessageMCPListTools,
     OpenAIResponseText,
+    OpenAISystemMessageParam,
+    OpenAIToolMessageParam,
+    OpenAIUserMessageParam,
+    ResponseGuardrailSpec,
+    Safety,
 )
-from llama_stack_api.safety import Safety
 
 
 async def convert_chat_choice_to_response_message(

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

@@ -6,8 +6,7 @@
 
 import asyncio
 
-from llama_stack_api.inference import OpenAIMessageParam
+from llama_stack_api import OpenAIMessageParam, Safety, SafetyViolation, ViolationLevel
-from llama_stack_api.safety import Safety, SafetyViolation, ViolationLevel
 
 from llama_stack.core.telemetry import tracing
 from llama_stack.log import get_logger

src/llama_stack/providers/inline/batches/reference/__init__.py

@@ -6,9 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.files import Files
+from llama_stack_api import Files, Inference, Models
-from llama_stack_api.inference import Inference
-from llama_stack_api.models import Models
 
 from llama_stack.core.datatypes import AccessRule, Api
 from llama_stack.providers.utils.kvstore import kvstore_impl

src/llama_stack/providers/inline/batches/reference/batches.py

@@ -13,22 +13,26 @@ import uuid
 from io import BytesIO
 from typing import Any, Literal
 
-from llama_stack_api.batches import Batches, BatchObject, ListBatchesResponse
+from llama_stack_api import (
-from llama_stack_api.common.errors import ConflictError, ResourceNotFoundError
+    Batches,
-from llama_stack_api.files import Files, OpenAIFilePurpose
+    BatchObject,
-from llama_stack_api.inference import (
+    ConflictError,
+    Files,
     Inference,
+    ListBatchesResponse,
+    Models,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletionRequestWithExtraBody,
     OpenAICompletionRequestWithExtraBody,
     OpenAIDeveloperMessageParam,
     OpenAIEmbeddingsRequestWithExtraBody,
+    OpenAIFilePurpose,
     OpenAIMessageParam,
     OpenAISystemMessageParam,
     OpenAIToolMessageParam,
     OpenAIUserMessageParam,
+    ResourceNotFoundError,
 )
-from llama_stack_api.models import Models
 from openai.types.batch import BatchError, Errors
 from pydantic import BaseModel
 

src/llama_stack/providers/inline/datasetio/localfs/datasetio.py

@@ -5,10 +5,7 @@
 # the root directory of this source tree.
 from typing import Any
 
-from llama_stack_api.common.responses import PaginatedResponse
+from llama_stack_api import Dataset, DatasetIO, DatasetsProtocolPrivate, PaginatedResponse
-from llama_stack_api.datasetio import DatasetIO
-from llama_stack_api.datasets import Dataset
-from llama_stack_api.datatypes import DatasetsProtocolPrivate
 
 from llama_stack.providers.utils.datasetio.url_utils import get_dataframe_from_uri
 from llama_stack.providers.utils.kvstore import kvstore_impl

src/llama_stack/providers/inline/eval/meta_reference/eval.py

@@ -6,21 +6,24 @@
 import json
 from typing import Any
 
-from llama_stack_api.agents import Agents
+from llama_stack_api import (
-from llama_stack_api.benchmarks import Benchmark
+    Agents,
-from llama_stack_api.common.job_types import Job, JobStatus
+    Benchmark,
-from llama_stack_api.datasetio import DatasetIO
+    BenchmarkConfig,
-from llama_stack_api.datasets import Datasets
+    BenchmarksProtocolPrivate,
-from llama_stack_api.datatypes import BenchmarksProtocolPrivate
+    DatasetIO,
-from llama_stack_api.eval import BenchmarkConfig, Eval, EvaluateResponse
+    Datasets,
-from llama_stack_api.inference import (
+    Eval,
+    EvaluateResponse,
     Inference,
+    Job,
+    JobStatus,
     OpenAIChatCompletionRequestWithExtraBody,
     OpenAICompletionRequestWithExtraBody,
     OpenAISystemMessageParam,
     OpenAIUserMessageParam,
+    Scoring,
 )
-from llama_stack_api.scoring import Scoring
 from tqdm import tqdm
 
 from llama_stack.providers.utils.common.data_schema_validator import ColumnName

src/llama_stack/providers/inline/files/localfs/files.py

@@ -10,15 +10,15 @@ from pathlib import Path
 from typing import Annotated
 
 from fastapi import Depends, File, Form, Response, UploadFile
-from llama_stack_api.common.errors import ResourceNotFoundError
+from llama_stack_api import (
-from llama_stack_api.common.responses import Order
-from llama_stack_api.files import (
     ExpiresAfter,
     Files,
     ListOpenAIFileResponse,
     OpenAIFileDeleteResponse,
     OpenAIFileObject,
     OpenAIFilePurpose,
+    Order,
+    ResourceNotFoundError,
 )
 
 from llama_stack.core.datatypes import AccessRule

src/llama_stack/providers/inline/inference/meta_reference/config.py

@@ -6,7 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.inference import QuantizationConfig
+from llama_stack_api import QuantizationConfig
 from pydantic import BaseModel, field_validator
 
 from llama_stack.providers.utils.inference import supported_inference_models

src/llama_stack/providers/inline/inference/meta_reference/generators.py

@@ -8,7 +8,7 @@ import math
 from typing import Optional
 
 import torch
-from llama_stack_api.inference import (
+from llama_stack_api import (
     GreedySamplingStrategy,
     JsonSchemaResponseFormat,
     OpenAIChatCompletionRequestWithExtraBody,

src/llama_stack/providers/inline/inference/meta_reference/inference.py

@@ -9,9 +9,11 @@ import time
 import uuid
 from collections.abc import AsyncIterator
 
-from llama_stack_api.datatypes import ModelsProtocolPrivate
+from llama_stack_api import (
-from llama_stack_api.inference import (
     InferenceProvider,
+    Model,
+    ModelsProtocolPrivate,
+    ModelType,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionChunk,
@@ -23,7 +25,6 @@ from llama_stack_api.inference import (
     OpenAIUserMessageParam,
     ToolChoice,
 )
-from llama_stack_api.models import Model, ModelType
 
 from llama_stack.log import get_logger
 from llama_stack.models.llama.datatypes import RawMessage, RawTextItem, ToolDefinition
@@ -375,7 +376,7 @@ class MetaReferenceInferenceImpl(
         # Convert tool calls to OpenAI format
         openai_tool_calls = None
         if decoded_message.tool_calls:
-            from llama_stack_api.inference import (
+            from llama_stack_api import (
                 OpenAIChatCompletionToolCall,
                 OpenAIChatCompletionToolCallFunction,
             )
@@ -440,7 +441,7 @@ class MetaReferenceInferenceImpl(
         params: OpenAIChatCompletionRequestWithExtraBody,
     ) -> AsyncIterator[OpenAIChatCompletionChunk]:
         """Stream chat completion chunks as they're generated."""
-        from llama_stack_api.inference import (
+        from llama_stack_api import (
             OpenAIChatCompletionChunk,
             OpenAIChatCompletionToolCall,
             OpenAIChatCompletionToolCallFunction,

src/llama_stack/providers/inline/inference/sentence_transformers/sentence_transformers.py

@@ -6,16 +6,17 @@
 
 from collections.abc import AsyncIterator
 
-from llama_stack_api.datatypes import ModelsProtocolPrivate
+from llama_stack_api import (
-from llama_stack_api.inference import (
     InferenceProvider,
+    Model,
+    ModelsProtocolPrivate,
+    ModelType,
     OpenAIChatCompletion,
     OpenAIChatCompletionChunk,
     OpenAIChatCompletionRequestWithExtraBody,
     OpenAICompletion,
     OpenAICompletionRequestWithExtraBody,
 )
-from llama_stack_api.models import Model, ModelType
 
 from llama_stack.log import get_logger
 from llama_stack.providers.utils.inference.embedding_mixin import (

src/llama_stack/providers/inline/post_training/common/validator.py

@@ -12,11 +12,7 @@
 
 from typing import Any
 
-from llama_stack_api.common.type_system import (
+from llama_stack_api import ChatCompletionInputType, DialogType, StringType
-    ChatCompletionInputType,
-    DialogType,
-    StringType,
-)
 
 from llama_stack.providers.utils.common.data_schema_validator import (
     ColumnName,

src/llama_stack/providers/inline/post_training/huggingface/post_training.py

@@ -6,11 +6,11 @@
 from enum import Enum
 from typing import Any
 
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.post_training import (
     AlgorithmConfig,
     Checkpoint,
+    DatasetIO,
+    Datasets,
     DPOAlignmentConfig,
     JobStatus,
     ListPostTrainingJobsResponse,

src/llama_stack/providers/inline/post_training/huggingface/recipes/finetune_single_device.py

@@ -12,11 +12,11 @@ from typing import Any
 
 import torch
 from datasets import Dataset
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.post_training import (
     Checkpoint,
     DataConfig,
+    DatasetIO,
+    Datasets,
     LoraFinetuningConfig,
     TrainingConfig,
 )

src/llama_stack/providers/inline/post_training/huggingface/recipes/finetune_single_device_dpo.py

@@ -11,10 +11,10 @@ from typing import Any
 
 import torch
 from datasets import Dataset
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.post_training import (
     Checkpoint,
+    DatasetIO,
+    Datasets,
     DPOAlignmentConfig,
     TrainingConfig,
 )

src/llama_stack/providers/inline/post_training/huggingface/utils.py

@@ -14,8 +14,7 @@ from typing import TYPE_CHECKING, Any, Protocol
 import psutil
 import torch
 from datasets import Dataset
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import Checkpoint, DatasetIO, TrainingConfig
-from llama_stack_api.post_training import Checkpoint, TrainingConfig
 from transformers import AutoConfig, AutoModelForCausalLM
 
 if TYPE_CHECKING:

src/llama_stack/providers/inline/post_training/torchtune/common/utils.py

@@ -13,7 +13,7 @@
 from collections.abc import Callable
 
 import torch
-from llama_stack_api.post_training import DatasetFormat
+from llama_stack_api import DatasetFormat
 from pydantic import BaseModel
 from torchtune.data._messages import InputOutputToMessages, ShareGPTToMessages
 from torchtune.models.llama3 import llama3_tokenizer

src/llama_stack/providers/inline/post_training/torchtune/post_training.py

@@ -6,11 +6,11 @@
 from enum import Enum
 from typing import Any
 
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.post_training import (
     AlgorithmConfig,
     Checkpoint,
+    DatasetIO,
+    Datasets,
     DPOAlignmentConfig,
     JobStatus,
     ListPostTrainingJobsResponse,

src/llama_stack/providers/inline/post_training/torchtune/recipes/lora_finetuning_single_device.py

@@ -12,14 +12,14 @@ from pathlib import Path
 from typing import Any
 
 import torch
-from llama_stack_api.common.training_types import PostTrainingMetric
+from llama_stack_api import (
-from llama_stack_api.datasetio import DatasetIO
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.post_training import (
     Checkpoint,
     DataConfig,
+    DatasetIO,
+    Datasets,
     LoraFinetuningConfig,
     OptimizerConfig,
+    PostTrainingMetric,
     QATFinetuningConfig,
     TrainingConfig,
 )

src/llama_stack/providers/inline/safety/code_scanner/code_scanner.py

@@ -10,16 +10,16 @@ from typing import TYPE_CHECKING, Any
 if TYPE_CHECKING:
     from codeshield.cs import CodeShieldScanResult
 
-from llama_stack_api.inference import OpenAIMessageParam
+from llama_stack_api import (
-from llama_stack_api.safety import (
     ModerationObject,
     ModerationObjectResults,
+    OpenAIMessageParam,
     RunShieldResponse,
     Safety,
     SafetyViolation,
+    Shield,
     ViolationLevel,
 )
-from llama_stack_api.shields import Shield
 
 from llama_stack.log import get_logger
 from llama_stack.providers.utils.inference.prompt_adapter import (

src/llama_stack/providers/inline/safety/llama_guard/llama_guard.py

@@ -9,23 +9,22 @@ import uuid
 from string import Template
 from typing import Any
 
-from llama_stack_api.common.content_types import ImageContentItem, TextContentItem
+from llama_stack_api import (
-from llama_stack_api.datatypes import ShieldsProtocolPrivate
+    ImageContentItem,
-from llama_stack_api.inference import (
     Inference,
+    ModerationObject,
+    ModerationObjectResults,
     OpenAIChatCompletionRequestWithExtraBody,
     OpenAIMessageParam,
     OpenAIUserMessageParam,
-)
-from llama_stack_api.safety import (
-    ModerationObject,
-    ModerationObjectResults,
     RunShieldResponse,
     Safety,
     SafetyViolation,
+    Shield,
+    ShieldsProtocolPrivate,
+    TextContentItem,
     ViolationLevel,
 )
-from llama_stack_api.shields import Shield
 
 from llama_stack.core.datatypes import Api
 from llama_stack.log import get_logger

src/llama_stack/providers/inline/safety/prompt_guard/prompt_guard.py

@@ -7,17 +7,17 @@
 from typing import Any
 
 import torch
-from llama_stack_api.datatypes import ShieldsProtocolPrivate
+from llama_stack_api import (
-from llama_stack_api.inference import OpenAIMessageParam
-from llama_stack_api.safety import (
     ModerationObject,
+    OpenAIMessageParam,
     RunShieldResponse,
     Safety,
     SafetyViolation,
+    Shield,
+    ShieldsProtocolPrivate,
     ShieldStore,
     ViolationLevel,
 )
-from llama_stack_api.shields import Shield
 from transformers import AutoModelForSequenceClassification, AutoTokenizer
 
 from llama_stack.core.utils.model_utils import model_local_dir

src/llama_stack/providers/inline/scoring/basic/scoring.py

@@ -5,16 +5,17 @@
 # the root directory of this source tree.
 from typing import Any
 
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
+    DatasetIO,
-from llama_stack_api.datatypes import ScoringFunctionsProtocolPrivate
+    Datasets,
-from llama_stack_api.scoring import (
     ScoreBatchResponse,
     ScoreResponse,
     Scoring,
+    ScoringFn,
+    ScoringFnParams,
+    ScoringFunctionsProtocolPrivate,
     ScoringResult,
 )
-from llama_stack_api.scoring_functions import ScoringFn, ScoringFnParams
 
 from llama_stack.core.datatypes import Api
 from llama_stack.providers.utils.common.data_schema_validator import (

src/llama_stack/providers/inline/scoring/basic/scoring_fn/docvqa_scoring_fn.py

@@ -8,8 +8,7 @@ import json
 import re
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/equality_scoring_fn.py

@@ -6,8 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/docvqa.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/equality.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/ifeval.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/regex_parser_math_response.py

@@ -4,9 +4,9 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
+    NumberType,
     RegexParserScoringFnParams,
     ScoringFn,
 )

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/regex_parser_multiple_choice_answer.py

@@ -4,9 +4,9 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
+    NumberType,
     RegexParserScoringFnParams,
     ScoringFn,
 )

src/llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/subset_of.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/ifeval_scoring_fn.py

@@ -6,8 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/regex_parser_math_response_scoring_fn.py

@@ -5,8 +5,7 @@
 # the root directory of this source tree.
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringFnParamsType, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams, ScoringFnParamsType
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/regex_parser_scoring_fn.py

@@ -6,8 +6,7 @@
 import re
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringFnParamsType, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams, ScoringFnParamsType
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/basic/scoring_fn/subset_of_scoring_fn.py

@@ -6,8 +6,7 @@
 
 from typing import Any
 
-from llama_stack_api.scoring import ScoringResultRow
+from llama_stack_api import ScoringFnParams, ScoringResultRow
-from llama_stack_api.scoring_functions import ScoringFnParams
 
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
 

src/llama_stack/providers/inline/scoring/braintrust/braintrust.py

@@ -17,17 +17,18 @@ from autoevals.ragas import (
     ContextRelevancy,
     Faithfulness,
 )
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
+    DatasetIO,
-from llama_stack_api.datatypes import ScoringFunctionsProtocolPrivate
+    Datasets,
-from llama_stack_api.scoring import (
     ScoreBatchResponse,
     ScoreResponse,
     Scoring,
+    ScoringFn,
+    ScoringFnParams,
+    ScoringFunctionsProtocolPrivate,
     ScoringResult,
     ScoringResultRow,
 )
-from llama_stack_api.scoring_functions import ScoringFn, ScoringFnParams
 from pydantic import BaseModel
 
 from llama_stack.core.datatypes import Api

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_correctness.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_relevancy.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/answer_similarity.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_entity_recall.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_precision.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_recall.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/context_relevancy.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/factuality.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/braintrust/scoring_fn/fn_defs/faithfulness.py

@@ -4,10 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.common.type_system import NumberType
+from llama_stack_api import (
-from llama_stack_api.scoring_functions import (
     AggregationFunctionType,
     BasicScoringFnParams,
+    NumberType,
     ScoringFn,
 )
 

src/llama_stack/providers/inline/scoring/llm_as_judge/scoring.py

@@ -5,17 +5,18 @@
 # the root directory of this source tree.
 from typing import Any
 
-from llama_stack_api.datasetio import DatasetIO
+from llama_stack_api import (
-from llama_stack_api.datasets import Datasets
+    DatasetIO,
-from llama_stack_api.datatypes import ScoringFunctionsProtocolPrivate
+    Datasets,
-from llama_stack_api.inference import Inference
+    Inference,
-from llama_stack_api.scoring import (
     ScoreBatchResponse,
     ScoreResponse,
     Scoring,
+    ScoringFn,
+    ScoringFnParams,
+    ScoringFunctionsProtocolPrivate,
     ScoringResult,
 )
-from llama_stack_api.scoring_functions import ScoringFn, ScoringFnParams
 
 from llama_stack.core.datatypes import Api
 from llama_stack.providers.utils.common.data_schema_validator import (