API Reference

Retrievers

Copy Markdown

Open in Claude

Open in ChatGPT

Open in Cursor

---

Copy Markdown

View as Markdown

Direct Retrieve

CompositeRetrievalResult retrievers().search(RetrieverSearchParamsparams, RequestOptionsrequestOptions = RequestOptions.none())

POST/api/v1/retrievers/retrieve

Retrieve data using specified pipelines without creating a persistent retriever.

ParametersExpand Collapse

RetrieverSearchParams params

Optional<String> organizationId

Optional<String> projectId

String query

The query to retrieve against.

minLength1

Optional<CompositeRetrievalMode> mode

The mode of composite retrieval.

Optional<List<RetrieverPipeline>> pipelines

The pipelines to use for retrieval.

Optional<String> description

A description of the retriever tool.

maxLength15000

Optional<String> name

A name for the retriever tool. Will default to the pipeline name if not provided.

maxLength3000

minLength1

String pipelineId

The ID of the pipeline this tool uses.

formatuuid

Optional<PresetRetrievalParams> presetRetrievalParameters

Parameters for retrieval configuration.

Optional<Double> alpha

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

maximum1

minimum0

Optional<String> className

Optional<Double> denseSimilarityCutoff

Minimum similarity score wrt query for retrieval

maximum1

minimum0

Optional<Long> denseSimilarityTopK

Number of nodes for dense retrieval.

maximum100

minimum1

Optional<Boolean> enableReranking

Enable reranking for retrieval

Optional<Long> filesTopK

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

maximum5

minimum1

Optional<Long> rerankTopN

Number of reranked nodes for returning.

maximum100

minimum1

Optional<RetrievalMode> retrievalMode

The retrieval mode for the query.

One of the following:

CHUNKS("chunks")

FILES_VIA_METADATA("files_via_metadata")

FILES_VIA_CONTENT("files_via_content")

AUTO_ROUTED("auto_routed")

DeprecatedOptional<Boolean> retrieveImageNodes

Whether to retrieve image nodes.

Optional<Boolean> retrievePageFigureNodes

Whether to retrieve page figure nodes.

Optional<Boolean> retrievePageScreenshotNodes

Whether to retrieve page screenshot nodes.

Optional<MetadataFilters> searchFilters

Metadata filters for vector stores.

List<Filter> filters

One of the following:

class MetadataFilter:

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

String key

Optional<Value> value

One of the following:

double

String

List<String>

List<double>

List<long>

Optional<Operator> operator

Vector store filter operator.

One of the following:

EQUALS("==")

GREATER(">")

LESS("<")

NOT_EQUALS("!=")

GREATER_OR_EQUALS(">=")

LESS_OR_EQUALS("<=")

IN("in")

NIN("nin")

ANY("any")

ALL("all")

TEXT_MATCH("text_match")

TEXT_MATCH_INSENSITIVE("text_match_insensitive")

CONTAINS("contains")

IS_EMPTY("is_empty")

MetadataFilters

Optional<Condition> condition

Vector store filter conditions to combine different filters.

One of the following:

AND("and")

OR("or")

NOT("not")

Optional<SearchFiltersInferenceSchema> searchFiltersInferenceSchema

JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference.

One of the following:

class UnionMember0:

List<JsonValue>

String

double

boolean

Optional<Long> sparseSimilarityTopK

Number of nodes for sparse retrieval.

maximum100

minimum1

Optional<ReRankConfig> rerankConfig

The rerank configuration for composite retrieval.

DeprecatedOptional<Long> rerankTopN

(use rerank_config.top_n instead) The number of nodes to retrieve after reranking over retrieved nodes from all retrieval tools.

ReturnsExpand Collapse

class CompositeRetrievalResult:

DeprecatedOptional<List<PageScreenshotNodeWithScore>> imageNodes

The image nodes retrieved by the pipeline for the given query. Deprecated - will soon be replaced with ‘page_screenshot_nodes’.

Node node

String fileId

The ID of the file that the page screenshot was taken from

formatuuid

long imageSize

The size of the image in bytes

minimum0

long pageIndex

The index of the page for which the screenshot is taken (0-indexed)

minimum0

Optional<Metadata> metadata

Metadata for the screenshot

double score

The score of the screenshot node

Optional<String> className

Optional<List<Node>> nodes

The retrieved nodes from the composite retrieval.

InnerNode node

String id

The ID of the retrieved node.

formatuuid

Optional<Long> endCharIdx

The end character index of the retrieved node in the document

String pipelineId

The ID of the pipeline this node was retrieved from.

formatuuid

String retrieverId

The ID of the retriever this node was retrieved from.

formatuuid

String retrieverPipelineName

The name of the retrieval pipeline this node was retrieved from.

Optional<Long> startCharIdx

The start character index of the retrieved node in the document

String text

The text of the retrieved node.

Optional<Metadata> metadata

Metadata associated with the retrieved node.

Optional<String> className

Optional<Double> score

Optional<List<PageFigureNodeWithScore>> pageFigureNodes

The page figure nodes retrieved by the pipeline for the given query.

Node node

double confidence

The confidence of the figure

maximum1

minimum0

String figureName

The name of the figure

long figureSize

The size of the figure in bytes

minimum0

String fileId

The ID of the file that the figure was taken from

formatuuid

long pageIndex

The index of the page for which the figure is taken (0-indexed)

minimum0

Optional<Boolean> isLikelyNoise

Whether the figure is likely to be noise

Optional<Metadata> metadata

Metadata for the figure

double score

The score of the figure node

Optional<String> className

Direct Retrieve

Java

HTTP

TypeScript

Python

Java

Go

CLI Tool

package com.llamacloud_prod.api.example;

import com.llamacloud_prod.api.client.LlamaCloudClient;
import com.llamacloud_prod.api.client.okhttp.LlamaCloudOkHttpClient;
import com.llamacloud_prod.api.models.retrievers.CompositeRetrievalResult;
import com.llamacloud_prod.api.models.retrievers.RetrieverSearchParams;

public final class Main {
    private Main() {}

    public static void main(String[] args) {
        LlamaCloudClient client = LlamaCloudOkHttpClient.fromEnv();

        RetrieverSearchParams params = RetrieverSearchParams.builder()
            .query("x")
            .build();
        CompositeRetrievalResult compositeRetrievalResult = client.retrievers().search(params);
    }
}

200 example

{
  "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"
    }
  ]
}

Returns Examples

200 example

{
  "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"
    }
  ]
}