## Direct Retrieve `retrievers.search(RetrieverSearchParams**kwargs) -> CompositeRetrievalResult` **post** `/api/v1/retrievers/retrieve` Retrieve data using specified pipelines without creating a persistent retriever. ### Parameters - `query: str` The query to retrieve against. - `organization_id: Optional[str]` - `project_id: Optional[str]` - `mode: Optional[CompositeRetrievalMode]` The mode of composite retrieval. - `"routing"` - `"full"` - `pipelines: Optional[Iterable[RetrieverPipelineParam]]` The pipelines to use for retrieval. - `description: Optional[str]` A description of the retriever tool. - `name: Optional[str]` A name for the retriever tool. Will default to the pipeline name if not provided. - `pipeline_id: str` The ID of the pipeline this tool uses. - `preset_retrieval_parameters: Optional[PresetRetrievalParams]` Parameters for retrieval configuration. - `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: Optional[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: Optional[RetrievalMode]` The retrieval mode for the query. - `"chunks"` - `"files_via_metadata"` - `"files_via_content"` - `"auto_routed"` - `retrieve_image_nodes: Optional[bool]` Whether to retrieve image nodes. - `retrieve_page_figure_nodes: Optional[bool]` Whether to retrieve page figure nodes. - `retrieve_page_screenshot_nodes: Optional[bool]` Whether to retrieve page screenshot nodes. - `search_filters: Optional[MetadataFilters]` Metadata filters for vector stores. - `filters: List[Filter]` - `class FilterMetadataFilter: …` Comprehensive metadata filter for vector stores to support more operators. Value uses Strict types, as int, float and str are compatible types and were all converted to string before. See: https://docs.pydantic.dev/latest/usage/types/#strict-types - `key: str` - `value: Union[float, str, List[str], 3 more]` - `float` - `str` - `List[str]` - `List[float]` - `List[int]` - `operator: Optional[Literal["==", ">", "<", 11 more]]` Vector store filter operator. - `"=="` - `">"` - `"<"` - `"!="` - `">="` - `"<="` - `"in"` - `"nin"` - `"any"` - `"all"` - `"text_match"` - `"text_match_insensitive"` - `"contains"` - `"is_empty"` - `class MetadataFilters: …` Metadata filters for vector stores. - `condition: Optional[Literal["and", "or", "not"]]` Vector store filter conditions to combine different filters. - `"and"` - `"or"` - `"not"` - `search_filters_inference_schema: Optional[Dict[str, Union[Dict[str, object], List[object], str, 3 more]]]` JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference. - `Dict[str, object]` - `List[object]` - `str` - `float` - `bool` - `sparse_similarity_top_k: Optional[int]` Number of nodes for sparse retrieval. - `rerank_config: Optional[ReRankConfigParam]` The rerank configuration for composite retrieval. - `top_n: Optional[int]` The number of nodes to retrieve after reranking over retrieved nodes from all retrieval tools. - `type: Optional[Literal["system_default", "llm", "cohere", 3 more]]` The type of reranker to use. - `"system_default"` - `"llm"` - `"cohere"` - `"bedrock"` - `"score"` - `"disabled"` - `rerank_top_n: Optional[int]` (use rerank_config.top_n instead) The number of nodes to retrieve after reranking over retrieved nodes from all retrieval tools. ### Returns - `class CompositeRetrievalResult: …` - `image_nodes: Optional[List[PageScreenshotNodeWithScore]]` The image nodes retrieved by the pipeline for the given query. Deprecated - will soon be replaced with 'page_screenshot_nodes'. - `node: Node` - `file_id: str` The ID of the file that the page screenshot was taken from - `image_size: int` The size of the image in bytes - `page_index: int` The index of the page for which the screenshot is taken (0-indexed) - `metadata: Optional[Dict[str, object]]` Metadata for the screenshot - `score: float` The score of the screenshot node - `class_name: Optional[str]` - `nodes: Optional[List[Node]]` The retrieved nodes from the composite retrieval. - `node: NodeNode` - `id: str` The ID of the retrieved node. - `end_char_idx: Optional[int]` The end character index of the retrieved node in the document - `pipeline_id: str` The ID of the pipeline this node was retrieved from. - `retriever_id: str` The ID of the retriever this node was retrieved from. - `retriever_pipeline_name: str` The name of the retrieval pipeline this node was retrieved from. - `start_char_idx: Optional[int]` The start character index of the retrieved node in the document - `text: str` The text of the retrieved node. - `metadata: Optional[Dict[str, object]]` Metadata associated with the retrieved node. - `class_name: Optional[str]` - `score: Optional[float]` - `page_figure_nodes: Optional[List[PageFigureNodeWithScore]]` The page figure nodes retrieved by the pipeline for the given query. - `node: Node` - `confidence: float` The confidence of the figure - `figure_name: str` The name of the figure - `figure_size: int` The size of the figure in bytes - `file_id: str` The ID of the file that the figure was taken from - `page_index: int` The index of the page for which the figure is taken (0-indexed) - `is_likely_noise: Optional[bool]` Whether the figure is likely to be noise - `metadata: Optional[Dict[str, object]]` Metadata for the figure - `score: float` The score of the figure node - `class_name: Optional[str]` ### Example ```python import os from llama_cloud import LlamaCloud client = LlamaCloud( api_key=os.environ.get("LLAMA_CLOUD_API_KEY"), # This is the default and can be omitted ) composite_retrieval_result = client.retrievers.search( query="x", ) print(composite_retrieval_result.image_nodes) ``` #### Response ```json { "image_nodes": [ { "node": { "file_id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e", "image_size": 0, "page_index": 0, "metadata": { "foo": "bar" } }, "score": 0, "class_name": "class_name" } ], "nodes": [ { "node": { "id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e", "end_char_idx": 0, "pipeline_id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e", "retriever_id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e", "retriever_pipeline_name": "retriever_pipeline_name", "start_char_idx": 0, "text": "text", "metadata": { "foo": "bar" } }, "class_name": "class_name", "score": 0 } ], "page_figure_nodes": [ { "node": { "confidence": 0, "figure_name": "figure_name", "figure_size": 0, "file_id": "182bd5e5-6e1a-4fe4-a799-aa6d9a6ab26e", "page_index": 0, "is_likely_noise": true, "metadata": { "foo": "bar" } }, "score": 0, "class_name": "class_name" } ] } ```