Skip to content
LlamaIndex API Docs
Get started
API Reference
Extraction

Extraction Agents

Create Extraction Agent
client.extraction.extractionAgents.create(ExtractionAgentCreateParams { config, data_schema, name, 2 more } params, RequestOptionsoptions?): ExtractAgent { id, config, data_schema, 5 more }
POST/api/v1/extraction/extraction-agents
List Extraction Agents
client.extraction.extractionAgents.list(ExtractionAgentListParams { include_default, organization_id, project_id } query?, RequestOptionsoptions?): ExtractionAgentListResponse { id, config, data_schema, 5 more }
GET/api/v1/extraction/extraction-agents
Get Extraction Agent
client.extraction.extractionAgents.get(stringextractionAgentID, RequestOptionsoptions?): ExtractAgent { id, config, data_schema, 5 more }
GET/api/v1/extraction/extraction-agents/{extraction_agent_id}
Delete Extraction Agent
client.extraction.extractionAgents.delete(stringextractionAgentID, RequestOptionsoptions?): ExtractionAgentDeleteResponse
DELETE/api/v1/extraction/extraction-agents/{extraction_agent_id}
Update Extraction Agent
client.extraction.extractionAgents.update(stringextractionAgentID, ExtractionAgentUpdateParams { config, data_schema } body, RequestOptionsoptions?): ExtractAgent { id, config, data_schema, 5 more }
PUT/api/v1/extraction/extraction-agents/{extraction_agent_id}
ModelsExpand Collapse
ExtractAgent { id, config, data_schema, 5 more }

Schema and configuration for an extraction agent.

id: string

The id of the extraction agent.

formatuuid
config: ExtractConfig { chunk_mode, citation_bbox, cite_sources, 13 more }

The configuration parameters for the extraction agent.

chunk_mode?: "PAGE" | "SECTION"

The mode to use for chunking the document.

Accepts one of the following:
"PAGE"
"SECTION"
Deprecatedcitation_bbox?: boolean

Whether to fetch citation bounding boxes for the extraction. Only available in PREMIUM mode. Deprecated: this is now synonymous with cite_sources.

cite_sources?: boolean

Whether to cite sources for the extraction.

confidence_scores?: boolean

Whether to fetch confidence scores for the extraction.

extract_model?: "openai-gpt-4-1" | "openai-gpt-4-1-mini" | "openai-gpt-4-1-nano" | 8 more | (string & {}) | null

The extract model to use for data extraction. If not provided, uses the default for the extraction mode.

Accepts one of the following:
"openai-gpt-4-1" | "openai-gpt-4-1-mini" | "openai-gpt-4-1-nano" | 8 more
"openai-gpt-4-1"
"openai-gpt-4-1-mini"
"openai-gpt-4-1-nano"
"openai-gpt-5"
"openai-gpt-5-mini"
"gemini-2.0-flash"
"gemini-2.5-flash"
"gemini-2.5-flash-lite"
"gemini-2.5-pro"
"openai-gpt-4o"
"openai-gpt-4o-mini"
(string & {})
extraction_mode?: "FAST" | "BALANCED" | "PREMIUM" | "MULTIMODAL"

The extraction mode specified (FAST, BALANCED, MULTIMODAL, PREMIUM).

Accepts one of the following:
"FAST"
"BALANCED"
"PREMIUM"
"MULTIMODAL"
extraction_target?: "PER_DOC" | "PER_PAGE" | "PER_TABLE_ROW"

The extraction target specified.

Accepts one of the following:
"PER_DOC"
"PER_PAGE"
"PER_TABLE_ROW"
high_resolution_mode?: boolean

Whether to use high resolution mode for the extraction.

invalidate_cache?: boolean

Whether to invalidate the cache for the extraction.

multimodal_fast_mode?: boolean

DEPRECATED: Whether to use fast mode for multimodal extraction.

num_pages_context?: number | null

Number of pages to pass as context on long document extraction.

minimum1
page_range?: string | null

Comma-separated list of page numbers or ranges to extract from (1-based, e.g., '1,3,5-7,9' or '1-3,8-10').

parse_model?: "openai-gpt-4o" | "openai-gpt-4o-mini" | "openai-gpt-4-1" | 23 more | null

Public model names.

Accepts one of the following:
"openai-gpt-4o"
"openai-gpt-4o-mini"
"openai-gpt-4-1"
"openai-gpt-4-1-mini"
"openai-gpt-4-1-nano"
"openai-gpt-5"
"openai-gpt-5-mini"
"openai-gpt-5-nano"
"openai-text-embedding-3-large"
"openai-text-embedding-3-small"
"openai-whisper-1"
"anthropic-sonnet-3.5"
"anthropic-sonnet-3.5-v2"
"anthropic-sonnet-3.7"
"anthropic-sonnet-4.0"
"anthropic-sonnet-4.5"
"anthropic-haiku-3.5"
"anthropic-haiku-4.5"
"gemini-2.5-flash"
"gemini-3.0-pro"
"gemini-2.5-pro"
"gemini-2.0-flash"
"gemini-2.0-flash-lite"
"gemini-2.5-flash-lite"
"gemini-1.5-flash"
"gemini-1.5-pro"
priority?: "low" | "medium" | "high" | "critical" | null

The priority for the request. This field may be ignored or overwritten depending on the organization tier.

Accepts one of the following:
"low"
"medium"
"high"
"critical"
system_prompt?: string | null

The system prompt to use for the extraction.

use_reasoning?: boolean

Whether to use reasoning for the extraction.

data_schema: Record<string, Record<string, unknown> | Array<unknown> | string | 2 more | null>

The schema of the data.

Accepts one of the following:
Record<string, unknown>
Array<unknown>
string
number
boolean
name: string

The name of the extraction agent.

project_id: string

The ID of the project that the extraction agent belongs to.

formatuuid
created_at?: string | null

The creation time of the extraction agent.

formatdate-time
custom_configuration?: "default" | null

Custom configuration type for the extraction agent. Currently supports 'default'.

updated_at?: string | null

The last update time of the extraction agent.

formatdate-time

Extraction AgentsSchema

Validate Extraction Schema
client.extraction.extractionAgents.schema.validateSchema(SchemaValidateSchemaParams { data_schema } body, RequestOptionsoptions?): SchemaValidateSchemaResponse { data_schema }
POST/api/v1/extraction/extraction-agents/schema/validation
Generate Extraction Schema
client.extraction.extractionAgents.schema.generateSchema(SchemaGenerateSchemaParams { organization_id, project_id, data_schema, 2 more } params, RequestOptionsoptions?): SchemaGenerateSchemaResponse { data_schema }
POST/api/v1/extraction/extraction-agents/schema/generate