llama-cloud 0.1.41__py3-none-any.whl → 1.0.0b4__py3-none-any.whl

llama_cloud/types/pipeline.py

@@ -1,73 +1,191 @@
-# This file was auto-generated by Fern from our API Definition.
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-import datetime as dt
-import typing
+from __future__ import annotations
 
-from ..core.datetime_utils import serialize_datetime
+from typing import Union, Optional
+from datetime import datetime
+from typing_extensions import Literal, Annotated, TypeAlias
+
+from pydantic import Field as FieldInfo
+
+from .._utils import PropertyInfo
+from .._models import BaseModel
 from .data_sink import DataSink
-from .embedding_model_config import EmbeddingModelConfig
-from .eval_execution_params import EvalExecutionParams
-from .llama_parse_parameters import LlamaParseParameters
-from .pipeline_configuration_hashes import PipelineConfigurationHashes
-from .pipeline_embedding_config import PipelineEmbeddingConfig
-from .pipeline_metadata_config import PipelineMetadataConfig
-from .pipeline_status import PipelineStatus
-from .pipeline_transform_config import PipelineTransformConfig
 from .pipeline_type import PipelineType
-from .preset_retrieval_params import PresetRetrievalParams
 from .sparse_model_config import SparseModelConfig
+from .auto_transform_config import AutoTransformConfig
+from .llama_parse_parameters import LlamaParseParameters
+from .cohere_embedding_config import CohereEmbeddingConfig
+from .gemini_embedding_config import GeminiEmbeddingConfig
+from .openai_embedding_config import OpenAIEmbeddingConfig
+from .bedrock_embedding_config import BedrockEmbeddingConfig
+from .pipeline_metadata_config import PipelineMetadataConfig
+from .vertex_ai_embedding_config import VertexAIEmbeddingConfig
+from .azure_openai_embedding_config import AzureOpenAIEmbeddingConfig
+from .advanced_mode_transform_config import AdvancedModeTransformConfig
+from .hugging_face_inference_api_embedding_config import HuggingFaceInferenceAPIEmbeddingConfig
 
-try:
-    import pydantic
-    if pydantic.__version__.startswith("1."):
-        raise ImportError
-    import pydantic.v1 as pydantic  # type: ignore
-except ImportError:
-    import pydantic  # type: ignore
+__all__ = [
+    "Pipeline",
+    "EmbeddingConfig",
+    "EmbeddingConfigManagedOpenAIEmbeddingConfig",
+    "EmbeddingConfigManagedOpenAIEmbeddingConfigComponent",
+    "ConfigHash",
+    "EmbeddingModelConfig",
+    "EmbeddingModelConfigEmbeddingConfig",
+    "TransformConfig",
+]
 
 
-class Pipeline(pydantic.BaseModel):
-    """
-    Schema for a pipeline.
-    """
+class EmbeddingConfigManagedOpenAIEmbeddingConfigComponent(BaseModel):
+    """Configuration for the Managed OpenAI embedding model."""
+
+    class_name: Optional[str] = None
+
+    embed_batch_size: Optional[int] = None
+    """The batch size for embedding calls."""
+
+    api_model_name: Optional[Literal["openai-text-embedding-3-small"]] = FieldInfo(alias="model_name", default=None)
+    """The name of the OpenAI embedding model."""
+
+    num_workers: Optional[int] = None
+    """The number of workers to use for async embedding calls."""
+
+
+class EmbeddingConfigManagedOpenAIEmbeddingConfig(BaseModel):
+    component: Optional[EmbeddingConfigManagedOpenAIEmbeddingConfigComponent] = None
+    """Configuration for the Managed OpenAI embedding model."""
+
+    type: Optional[Literal["MANAGED_OPENAI_EMBEDDING"]] = None
+    """Type of the embedding model."""
+
+
+EmbeddingConfig: TypeAlias = Annotated[
+    Union[
+        EmbeddingConfigManagedOpenAIEmbeddingConfig,
+        AzureOpenAIEmbeddingConfig,
+        CohereEmbeddingConfig,
+        GeminiEmbeddingConfig,
+        HuggingFaceInferenceAPIEmbeddingConfig,
+        OpenAIEmbeddingConfig,
+        VertexAIEmbeddingConfig,
+        BedrockEmbeddingConfig,
+    ],
+    PropertyInfo(discriminator="type"),
+]
+
+
+class ConfigHash(BaseModel):
+    """Hashes for the configuration of a pipeline."""
+
+    embedding_config_hash: Optional[str] = None
+    """Hash of the embedding config."""
+
+    parsing_config_hash: Optional[str] = None
+    """Hash of the llama parse parameters."""
+
+    transform_config_hash: Optional[str] = None
+    """Hash of the transform config."""
+
+
+EmbeddingModelConfigEmbeddingConfig: TypeAlias = Annotated[
+    Union[
+        AzureOpenAIEmbeddingConfig,
+        CohereEmbeddingConfig,
+        GeminiEmbeddingConfig,
+        HuggingFaceInferenceAPIEmbeddingConfig,
+        OpenAIEmbeddingConfig,
+        VertexAIEmbeddingConfig,
+        BedrockEmbeddingConfig,
+    ],
+    PropertyInfo(discriminator="type"),
+]
+
+
+class EmbeddingModelConfig(BaseModel):
+    """Schema for an embedding model config."""
+
+    id: str
+    """Unique identifier"""
+
+    embedding_config: EmbeddingModelConfigEmbeddingConfig
+    """The embedding configuration for the embedding model config."""
 
-    id: str = pydantic.Field(description="Unique identifier")
-    created_at: typing.Optional[dt.datetime]
-    updated_at: typing.Optional[dt.datetime]
     name: str
+    """The name of the embedding model config."""
+
     project_id: str
-    embedding_model_config_id: typing.Optional[str]
-    embedding_model_config: typing.Optional[EmbeddingModelConfig]
-    pipeline_type: typing.Optional[PipelineType] = pydantic.Field(
-        description="Type of pipeline. Either PLAYGROUND or MANAGED."
-    )
-    managed_pipeline_id: typing.Optional[str]
-    embedding_config: PipelineEmbeddingConfig
-    sparse_model_config: typing.Optional[SparseModelConfig]
-    config_hash: typing.Optional[PipelineConfigurationHashes]
-    transform_config: typing.Optional[PipelineTransformConfig] = pydantic.Field(
-        description="Configuration for the transformation."
-    )
-    preset_retrieval_parameters: typing.Optional[PresetRetrievalParams] = pydantic.Field(
-        description="Preset retrieval parameters for the pipeline."
-    )
-    eval_parameters: typing.Optional[EvalExecutionParams] = pydantic.Field(
-        description="Eval parameters for the pipeline."
-    )
-    llama_parse_parameters: typing.Optional[LlamaParseParameters]
-    data_sink: typing.Optional[DataSink]
-    status: typing.Optional[PipelineStatus]
-    metadata_config: typing.Optional[PipelineMetadataConfig]
-
-    def json(self, **kwargs: typing.Any) -> str:
-        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
-        return super().json(**kwargs_with_defaults)
-
-    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
-        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
-        return super().dict(**kwargs_with_defaults)
-
-    class Config:
-        frozen = True
-        smart_union = True
-        json_encoders = {dt.datetime: serialize_datetime}
+
+    created_at: Optional[datetime] = None
+    """Creation datetime"""
+
+    updated_at: Optional[datetime] = None
+    """Update datetime"""
+
+
+TransformConfig: TypeAlias = Union[AutoTransformConfig, AdvancedModeTransformConfig]
+
+
+class Pipeline(BaseModel):
+    """Schema for a pipeline."""
+
+    id: str
+    """Unique identifier"""
+
+    embedding_config: EmbeddingConfig
+
+    name: str
+
+    project_id: str
+
+    config_hash: Optional[ConfigHash] = None
+    """Hashes for the configuration of a pipeline."""
+
+    created_at: Optional[datetime] = None
+    """Creation datetime"""
+
+    data_sink: Optional[DataSink] = None
+    """Schema for a data sink."""
+
+    embedding_model_config: Optional[EmbeddingModelConfig] = None
+    """Schema for an embedding model config."""
+
+    embedding_model_config_id: Optional[str] = None
+    """The ID of the EmbeddingModelConfig this pipeline is using."""
+
+    llama_parse_parameters: Optional[LlamaParseParameters] = None
+    """
+    Settings that can be configured for how to use LlamaParse to parse files within
+    a LlamaCloud pipeline.
+    """
+
+    managed_pipeline_id: Optional[str] = None
+    """The ID of the ManagedPipeline this playground pipeline is linked to."""
+
+    metadata_config: Optional[PipelineMetadataConfig] = None
+    """Metadata configuration for the pipeline."""
+
+    pipeline_type: Optional[PipelineType] = None
+    """Type of pipeline. Either PLAYGROUND or MANAGED."""
+
+    preset_retrieval_parameters: Optional["PresetRetrievalParams"] = None
+    """Preset retrieval parameters for the pipeline."""
+
+    sparse_model_config: Optional[SparseModelConfig] = None
+    """Configuration for sparse embedding models used in hybrid search.
+
+    This allows users to choose between Splade and BM25 models for sparse retrieval
+    in managed data sinks.
+    """
+
+    status: Optional[Literal["CREATED", "DELETING"]] = None
+    """Status of the pipeline."""
+
+    transform_config: Optional[TransformConfig] = None
+    """Configuration for the transformation."""
+
+    updated_at: Optional[datetime] = None
+    """Update datetime"""
+
+
+from .preset_retrieval_params import PresetRetrievalParams

llama_cloud/types/pipeline_create_params.py

@@ -0,0 +1,95 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Optional
+from typing_extensions import Required, TypeAlias, TypedDict
+
+from .pipeline_type import PipelineType
+from .data_sink_create_param import DataSinkCreateParam
+from .sparse_model_config_param import SparseModelConfigParam
+from .auto_transform_config_param import AutoTransformConfigParam
+from .llama_parse_parameters_param import LlamaParseParametersParam
+from .cohere_embedding_config_param import CohereEmbeddingConfigParam
+from .gemini_embedding_config_param import GeminiEmbeddingConfigParam
+from .openai_embedding_config_param import OpenAIEmbeddingConfigParam
+from .bedrock_embedding_config_param import BedrockEmbeddingConfigParam
+from .pipeline_metadata_config_param import PipelineMetadataConfigParam
+from .vertex_ai_embedding_config_param import VertexAIEmbeddingConfigParam
+from .azure_openai_embedding_config_param import AzureOpenAIEmbeddingConfigParam
+from .advanced_mode_transform_config_param import AdvancedModeTransformConfigParam
+from .hugging_face_inference_api_embedding_config_param import HuggingFaceInferenceAPIEmbeddingConfigParam
+
+__all__ = ["PipelineCreateParams", "EmbeddingConfig", "TransformConfig"]
+
+
+class PipelineCreateParams(TypedDict, total=False):
+    name: Required[str]
+
+    organization_id: Optional[str]
+
+    project_id: Optional[str]
+
+    data_sink: Optional[DataSinkCreateParam]
+    """Schema for creating a data sink."""
+
+    data_sink_id: Optional[str]
+    """Data sink ID.
+
+    When provided instead of data_sink, the data sink will be looked up by ID.
+    """
+
+    embedding_config: Optional[EmbeddingConfig]
+
+    embedding_model_config_id: Optional[str]
+    """Embedding model config ID.
+
+    When provided instead of embedding_config, the embedding model config will be
+    looked up by ID.
+    """
+
+    llama_parse_parameters: LlamaParseParametersParam
+    """
+    Settings that can be configured for how to use LlamaParse to parse files within
+    a LlamaCloud pipeline.
+    """
+
+    managed_pipeline_id: Optional[str]
+    """The ID of the ManagedPipeline this playground pipeline is linked to."""
+
+    metadata_config: Optional[PipelineMetadataConfigParam]
+    """Metadata configuration for the pipeline."""
+
+    pipeline_type: PipelineType
+    """Type of pipeline. Either PLAYGROUND or MANAGED."""
+
+    preset_retrieval_parameters: "PresetRetrievalParamsParam"
+    """Preset retrieval parameters for the pipeline."""
+
+    sparse_model_config: Optional[SparseModelConfigParam]
+    """Configuration for sparse embedding models used in hybrid search.
+
+    This allows users to choose between Splade and BM25 models for sparse retrieval
+    in managed data sinks.
+    """
+
+    status: Optional[str]
+    """Status of the pipeline deployment."""
+
+    transform_config: Optional[TransformConfig]
+    """Configuration for the transformation."""
+
+
+EmbeddingConfig: TypeAlias = Union[
+    AzureOpenAIEmbeddingConfigParam,
+    CohereEmbeddingConfigParam,
+    GeminiEmbeddingConfigParam,
+    HuggingFaceInferenceAPIEmbeddingConfigParam,
+    OpenAIEmbeddingConfigParam,
+    VertexAIEmbeddingConfigParam,
+    BedrockEmbeddingConfigParam,
+]
+
+TransformConfig: TypeAlias = Union[AutoTransformConfigParam, AdvancedModeTransformConfigParam]
+
+from .preset_retrieval_params_param import PresetRetrievalParamsParam

llama_cloud/types/pipeline_get_status_params.py

@@ -0,0 +1,12 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Optional
+from typing_extensions import TypedDict
+
+__all__ = ["PipelineGetStatusParams"]
+
+
+class PipelineGetStatusParams(TypedDict, total=False):
+    full_details: Optional[bool]

llama_cloud/types/pipeline_list_params.py

@@ -0,0 +1,23 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Optional
+from typing_extensions import TypedDict
+
+from .pipeline_type import PipelineType
+
+__all__ = ["PipelineListParams"]
+
+
+class PipelineListParams(TypedDict, total=False):
+    organization_id: Optional[str]
+
+    pipeline_name: Optional[str]
+
+    pipeline_type: Optional[PipelineType]
+    """Enum for representing the type of a pipeline"""
+
+    project_id: Optional[str]
+
+    project_name: Optional[str]

llama_cloud/types/pipeline_list_response.py

@@ -0,0 +1,12 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List
+from typing_extensions import TypeAlias
+
+__all__ = ["PipelineListResponse"]
+
+from .pipeline import Pipeline
+
+PipelineListResponse: TypeAlias = List[Pipeline]

llama_cloud/types/pipeline_metadata_config.py

@@ -1,36 +1,15 @@
-# This file was auto-generated by Fern from our API Definition.
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-import datetime as dt
-import typing
+from typing import List, Optional
 
-from ..core.datetime_utils import serialize_datetime
+from .._models import BaseModel
 
-try:
-    import pydantic
-    if pydantic.__version__.startswith("1."):
-        raise ImportError
-    import pydantic.v1 as pydantic  # type: ignore
-except ImportError:
-    import pydantic  # type: ignore
+__all__ = ["PipelineMetadataConfig"]
 
 
-class PipelineMetadataConfig(pydantic.BaseModel):
-    excluded_embed_metadata_keys: typing.Optional[typing.List[str]] = pydantic.Field(
-        description="List of metadata keys to exclude from embeddings"
-    )
-    excluded_llm_metadata_keys: typing.Optional[typing.List[str]] = pydantic.Field(
-        description="List of metadata keys to exclude from LLM during retrieval"
-    )
+class PipelineMetadataConfig(BaseModel):
+    excluded_embed_metadata_keys: Optional[List[str]] = None
+    """List of metadata keys to exclude from embeddings"""
 
-    def json(self, **kwargs: typing.Any) -> str:
-        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
-        return super().json(**kwargs_with_defaults)
-
-    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
-        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
-        return super().dict(**kwargs_with_defaults)
-
-    class Config:
-        frozen = True
-        smart_union = True
-        json_encoders = {dt.datetime: serialize_datetime}
+    excluded_llm_metadata_keys: Optional[List[str]] = None
+    """List of metadata keys to exclude from LLM during retrieval"""

llama_cloud/types/pipeline_metadata_config_param.py

@@ -0,0 +1,17 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing_extensions import TypedDict
+
+from .._types import SequenceNotStr
+
+__all__ = ["PipelineMetadataConfigParam"]
+
+
+class PipelineMetadataConfigParam(TypedDict, total=False):
+    excluded_embed_metadata_keys: SequenceNotStr[str]
+    """List of metadata keys to exclude from embeddings"""
+
+    excluded_llm_metadata_keys: SequenceNotStr[str]
+    """List of metadata keys to exclude from LLM during retrieval"""

llama_cloud/types/pipeline_retrieve_params.py

@@ -0,0 +1,74 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Dict, Union, Iterable, Optional
+from typing_extensions import Required, TypedDict
+
+from .retrieval_mode import RetrievalMode
+
+__all__ = ["PipelineRetrieveParams"]
+
+
+class PipelineRetrieveParams(TypedDict, total=False):
+    query: Required[str]
+    """The query to retrieve against."""
+
+    organization_id: Optional[str]
+
+    project_id: Optional[str]
+
+    alpha: Optional[float]
+    """
+    Alpha value for hybrid retrieval to determine the weights between dense and
+    sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.
+    """
+
+    class_name: str
+
+    dense_similarity_cutoff: Optional[float]
+    """Minimum similarity score wrt query for retrieval"""
+
+    dense_similarity_top_k: Optional[int]
+    """Number of nodes for dense retrieval."""
+
+    enable_reranking: Optional[bool]
+    """Enable reranking for retrieval"""
+
+    files_top_k: Optional[int]
+    """
+    Number of files to retrieve (only for retrieval mode files_via_metadata and
+    files_via_content).
+    """
+
+    rerank_top_n: Optional[int]
+    """Number of reranked nodes for returning."""
+
+    retrieval_mode: RetrievalMode
+    """The retrieval mode for the query."""
+
+    retrieve_image_nodes: bool
+    """Whether to retrieve image nodes."""
+
+    retrieve_page_figure_nodes: bool
+    """Whether to retrieve page figure nodes."""
+
+    retrieve_page_screenshot_nodes: bool
+    """Whether to retrieve page screenshot nodes."""
+
+    search_filters: Optional["MetadataFiltersParam"]
+    """Metadata filters for vector stores."""
+
+    search_filters_inference_schema: Optional[
+        Dict[str, Union[Dict[str, object], Iterable[object], str, float, bool, None]]
+    ]
+    """JSON Schema that will be used to infer search_filters.
+
+    Omit or leave as null to skip inference.
+    """
+
+    sparse_similarity_top_k: Optional[int]
+    """Number of nodes for sparse retrieval."""
+
+
+from .metadata_filters_param import MetadataFiltersParam

llama_cloud/types/pipeline_retrieve_response.py

@@ -0,0 +1,63 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Dict, List, Optional
+
+from .._models import BaseModel
+from .pipelines.text_node import TextNode
+from .page_figure_node_with_score import PageFigureNodeWithScore
+from .page_screenshot_node_with_score import PageScreenshotNodeWithScore
+
+__all__ = ["PipelineRetrieveResponse", "RetrievalNode"]
+
+
+class RetrievalNode(BaseModel):
+    """
+    Same as NodeWithScore but type for node is a TextNode instead of BaseNode.
+    FastAPI doesn't accept abstract classes like BaseNode.
+    """
+
+    node: TextNode
+    """Provided for backward compatibility.
+
+    Note: we keep the field with the typo "seperator" to maintain backward
+    compatibility for serialized objects.
+    """
+
+    class_name: Optional[str] = None
+
+    score: Optional[float] = None
+
+
+class PipelineRetrieveResponse(BaseModel):
+    """Schema for the result of an retrieval execution."""
+
+    pipeline_id: str
+    """The ID of the pipeline that the query was retrieved against."""
+
+    retrieval_nodes: List[RetrievalNode]
+    """The nodes retrieved by the pipeline for the given query."""
+
+    class_name: Optional[str] = None
+
+    image_nodes: Optional[List[PageScreenshotNodeWithScore]] = None
+    """The image nodes retrieved by the pipeline for the given query.
+
+    Deprecated - will soon be replaced with 'page_screenshot_nodes'.
+    """
+
+    inferred_search_filters: Optional["MetadataFilters"] = None
+    """Metadata filters for vector stores."""
+
+    metadata: Optional[Dict[str, str]] = None
+    """Metadata associated with the retrieval execution"""
+
+    page_figure_nodes: Optional[List[PageFigureNodeWithScore]] = None
+    """The page figure nodes retrieved by the pipeline for the given query."""
+
+    retrieval_latency: Optional[Dict[str, float]] = None
+    """The end-to-end latency for retrieval and reranking."""
+
+
+from .metadata_filters import MetadataFilters

llama_cloud/types/pipeline_type.py

@@ -1,21 +1,7 @@
-# This file was auto-generated by Fern from our API Definition.
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-import enum
-import typing
+from typing_extensions import Literal, TypeAlias
 
-T_Result = typing.TypeVar("T_Result")
+__all__ = ["PipelineType"]
 
-
-class PipelineType(str, enum.Enum):
-    """
-    Enum for representing the type of a pipeline
-    """
-
-    PLAYGROUND = "PLAYGROUND"
-    MANAGED = "MANAGED"
-
-    def visit(self, playground: typing.Callable[[], T_Result], managed: typing.Callable[[], T_Result]) -> T_Result:
-        if self is PipelineType.PLAYGROUND:
-            return playground()
-        if self is PipelineType.MANAGED:
-            return managed()
+PipelineType: TypeAlias = Literal["PLAYGROUND", "MANAGED"]