The LlamaCloud Index API provides a fully automated document ingestion pipeline with built-in retrieval capabilities. It allows developers to create and manage indexes, upload documents for automatic parsing and embedding, and perform retrieval queries against indexed content. The API supports customizable pipeline configurations and integrates with various vector stores, making it suitable for building production RAG applications without managing infrastructure.
openapi: 3.1.0
info:
title: LlamaIndex LlamaCloud Index API
description: >-
The LlamaCloud Index API provides a fully automated document ingestion
pipeline with built-in retrieval capabilities. It allows developers to
create and manage indexes (pipelines), upload documents for automatic
parsing and embedding, and perform retrieval queries against indexed
content. The API supports customizable pipeline configurations and
integrates with various vector stores, making it suitable for building
production RAG applications without managing infrastructure.
version: '1.0'
contact:
name: LlamaIndex Support
url: https://www.llamaindex.ai/contact
termsOfService: https://www.llamaindex.ai/terms-of-service
externalDocs:
description: Index API and Clients Guide
url: https://developers.llamaindex.ai/python/cloud/llamacloud/guides/api_sdk/
servers:
- url: https://api.cloud.llamaindex.ai/api/v1
description: US Production Server
- url: https://api.cloud.llamaindex.eu/api/v1
description: EU Production Server
tags:
- name: Index Data Sources
description: >-
Connect and manage external data sources for automatic document
ingestion into an index.
- name: Index Files
description: >-
Manage files within an index, including upload, listing, and status
monitoring.
- name: Index Retrieval
description: >-
Query indexed content using configurable retrieval modes and parameters.
- name: Index Status
description: >-
Monitor the processing status and health of an index.
- name: Indexes
description: >-
Create and manage indexes that serve as automated document ingestion
pipelines with retrieval capabilities.
security:
- bearerAuth: []
paths:
/pipelines:
get:
operationId: listIndexes
summary: List indexes
description: >-
Retrieve a list of all indexes (pipelines) for the specified project.
Each index represents an automated document ingestion pipeline with
retrieval capabilities.
tags:
- Indexes
parameters:
- name: project_id
in: query
description: >-
Filter indexes by project identifier.
required: true
schema:
type: string
responses:
'200':
description: Successfully retrieved list of indexes
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/Index'
'401':
description: Unauthorized - invalid or missing API key
post:
operationId: createIndex
summary: Create an index
description: >-
Create a new index with specified configuration for document ingestion,
parsing, embedding, and vector storage. The index automatically handles
processing uploaded documents into a queryable format.
tags:
- Indexes
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/CreateIndexRequest'
responses:
'201':
description: Index created successfully
content:
application/json:
schema:
$ref: '#/components/schemas/Index'
'400':
description: Bad request - invalid index configuration
'401':
description: Unauthorized - invalid or missing API key
/pipelines/{pipelineId}:
get:
operationId: getIndex
summary: Get an index
description: >-
Retrieve detailed information about a specific index including its
configuration, processing status, and file counts.
tags:
- Indexes
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Successfully retrieved index details
content:
application/json:
schema:
$ref: '#/components/schemas/Index'
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
delete:
operationId: deleteIndex
summary: Delete an index
description: >-
Delete an index and its associated processed data.
tags:
- Indexes
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Index deleted successfully
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
/pipelines/{pipelineId}/sync:
post:
operationId: syncIndex
summary: Sync an index
description: >-
Trigger an incremental sync of the index, processing any new or
modified documents from connected data sources and uploaded files.
tags:
- Indexes
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Index sync initiated successfully
content:
application/json:
schema:
$ref: '#/components/schemas/IndexJob'
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
/pipelines/{pipelineId}/files:
get:
operationId: listIndexFiles
summary: List index files
description: >-
Retrieve a list of files in the index along with their processing
status. Use this endpoint to monitor which files have been
successfully indexed.
tags:
- Index Files
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Successfully retrieved list of index files
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/IndexFile'
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
put:
operationId: addFilesToIndex
summary: Add files to an index
description: >-
Add previously uploaded files to the index for automatic processing.
Files can include optional custom metadata that is preserved through
the indexing pipeline.
tags:
- Index Files
parameters:
- $ref: '#/components/parameters/pipelineId'
requestBody:
required: true
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/IndexFileInput'
responses:
'200':
description: Files added to index successfully
'400':
description: Bad request - invalid file references
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
/pipelines/{pipelineId}/files/count:
get:
operationId: getIndexFileCounts
summary: Get index file counts
description: >-
Retrieve file processing counts for an index, broken down by status.
Use this endpoint to determine when an index is ready to query by
checking that files have been successfully processed.
tags:
- Index Status
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Successfully retrieved file counts
content:
application/json:
schema:
$ref: '#/components/schemas/IndexFileCounts'
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
/pipelines/{pipelineId}/data-sources:
get:
operationId: listIndexDataSources
summary: List index data sources
description: >-
Retrieve data sources connected to a specific index for automatic
document ingestion.
tags:
- Index Data Sources
parameters:
- $ref: '#/components/parameters/pipelineId'
responses:
'200':
description: Successfully retrieved list of data sources
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/DataSource'
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
put:
operationId: addDataSourcesToIndex
summary: Add data sources to an index
description: >-
Connect external data sources to an index for automatic document
ingestion. The index will sync content from these sources based
on the configured schedule.
tags:
- Index Data Sources
parameters:
- $ref: '#/components/parameters/pipelineId'
requestBody:
required: true
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/DataSourceInput'
responses:
'200':
description: Data sources added to index successfully
'400':
description: Bad request - invalid data source configuration
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
/pipelines/{pipelineId}/retrieve:
post:
operationId: retrieveFromIndex
summary: Retrieve from index
description: >-
Execute a retrieval query against the indexed content. Supports
multiple retrieval modes (fast, accurate, advanced) and configurable
parameters for controlling the number of results and re-ranking
behavior.
tags:
- Index Retrieval
parameters:
- $ref: '#/components/parameters/pipelineId'
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/RetrievalRequest'
responses:
'200':
description: Retrieval results returned successfully
content:
application/json:
schema:
$ref: '#/components/schemas/RetrievalResponse'
'400':
description: Bad request - invalid query parameters
'401':
description: Unauthorized - invalid or missing API key
'404':
description: Index not found
components:
securitySchemes:
bearerAuth:
type: http
scheme: bearer
bearerFormat: API Key
description: >-
LlamaCloud API key obtained from the LlamaCloud dashboard. Include as
a Bearer token in the Authorization header.
parameters:
pipelineId:
name: pipelineId
in: path
description: >-
Unique identifier of the index (pipeline).
required: true
schema:
type: string
schemas:
Index:
type: object
description: >-
An index representing an automated document ingestion pipeline with
built-in retrieval capabilities.
properties:
id:
type: string
description: >-
Unique identifier of the index.
name:
type: string
description: >-
Human-readable name of the index.
project_id:
type: string
description: >-
Identifier of the project the index belongs to.
status:
type: string
enum:
- active
- syncing
- error
- idle
description: >-
Current status of the index.
embedding_model:
type: string
description: >-
Name of the embedding model used for vectorization.
sync_interval:
type: integer
description: >-
Scheduled sync frequency in seconds, if configured.
created_at:
type: string
format: date-time
description: >-
Timestamp when the index was created.
updated_at:
type: string
format: date-time
description: >-
Timestamp when the index was last updated.
CreateIndexRequest:
type: object
description: >-
Request body for creating a new index.
required:
- name
- project_id
properties:
name:
type: string
description: >-
Human-readable name for the new index.
project_id:
type: string
description: >-
Identifier of the project to create the index in.
embedding_model:
type: string
description: >-
Name of the embedding model to use for vectorization.
sync_interval:
type: integer
description: >-
Optional scheduled sync frequency in seconds.
IndexFile:
type: object
description: >-
A file within an index, including its processing status.
properties:
file_id:
type: string
description: >-
Unique identifier of the file.
name:
type: string
description: >-
Original file name.
status:
type: string
enum:
- pending
- processing
- completed
- error
description: >-
Processing status of the file within the index.
metadata:
type: object
additionalProperties: true
description: >-
Custom metadata associated with the file.
IndexFileInput:
type: object
description: >-
Input for adding a file to an index.
required:
- file_id
properties:
file_id:
type: string
description: >-
Identifier of the file to add to the index.
metadata:
type: object
additionalProperties: true
description: >-
Optional custom metadata to associate with the file.
IndexFileCounts:
type: object
description: >-
File processing counts for an index, broken down by status.
properties:
total:
type: integer
description: >-
Total number of files in the index.
pending:
type: integer
description: >-
Number of files pending processing.
processing:
type: integer
description: >-
Number of files currently being processed.
completed:
type: integer
description: >-
Number of files successfully processed.
error:
type: integer
description: >-
Number of files that encountered processing errors.
IndexJob:
type: object
description: >-
A sync job for an index tracking the processing of documents.
properties:
id:
type: string
description: >-
Unique identifier of the job.
pipeline_id:
type: string
description: >-
Identifier of the index the job belongs to.
status:
type: string
enum:
- pending
- running
- completed
- failed
description: >-
Current status of the job.
created_at:
type: string
format: date-time
description: >-
Timestamp when the job was created.
completed_at:
type: string
format: date-time
description: >-
Timestamp when the job completed, if applicable.
DataSource:
type: object
description: >-
A data source connected to an index for automatic document ingestion.
properties:
id:
type: string
description: >-
Unique identifier of the data source.
name:
type: string
description: >-
Human-readable name of the data source.
source_type:
type: string
description: >-
Type of the data source.
config:
type: object
additionalProperties: true
description: >-
Configuration specific to the data source type.
DataSourceInput:
type: object
description: >-
Input for connecting a data source to an index.
required:
- name
- source_type
properties:
name:
type: string
description: >-
Human-readable name for the data source.
source_type:
type: string
description: >-
Type of the data source.
config:
type: object
additionalProperties: true
description: >-
Configuration specific to the data source type.
RetrievalRequest:
type: object
description: >-
Request body for running a retrieval query against an index.
required:
- query
properties:
query:
type: string
description: >-
The search query to run against the indexed content.
mode:
type: string
enum:
- fast
- accurate
- advanced
description: >-
Retrieval mode controlling the trade-off between speed and
accuracy.
top_k:
type: integer
minimum: 1
description: >-
Maximum number of results to return after re-ranking.
RetrievalResponse:
type: object
description: >-
Response containing retrieval results from an index query.
properties:
results:
type: array
description: >-
List of retrieval results ranked by relevance.
items:
$ref: '#/components/schemas/RetrievalResult'
RetrievalResult:
type: object
description: >-
A single retrieval result with matched content and metadata.
properties:
text:
type: string
description: >-
The retrieved text content.
score:
type: number
format: float
description: >-
Relevance score for the result.
metadata:
type: object
additionalProperties: true
description: >-
Metadata associated with the retrieved content.