The configuration parameters for the extraction agent.
chunk_mode: optional "PAGE" or "SECTION"
The mode to use for chunking the document.
One of the following:
"PAGE"
"SECTION"
Deprecatedcitation_bbox: optional 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: optional boolean
Whether to cite sources for the extraction.
confidence_scores: optional boolean
Whether to fetch confidence scores for the extraction.
extract_model: optional "gemini-2.0-flash" or "gemini-2.5-flash" or "gemini-2.5-flash-lite" or 8 more or string
The extract model to use for data extraction. If not provided, uses the default for the extraction mode.
One of the following:
ExtractModels = "gemini-2.0-flash" or "gemini-2.5-flash" or "gemini-2.5-flash-lite" or 8 more
Extract model options.
One of the following:
"gemini-2.0-flash"
"gemini-2.5-flash"
"gemini-2.5-flash-lite"
"gemini-2.5-pro"
"openai-gpt-4-1"
"openai-gpt-4-1-mini"
"openai-gpt-4-1-nano"
"openai-gpt-4o"
"openai-gpt-4o-mini"
"openai-gpt-5"
"openai-gpt-5-mini"
string
extraction_mode: optional "BALANCED" or "FAST" or "MULTIMODAL" or "PREMIUM"
The extraction mode specified (FAST, BALANCED, MULTIMODAL, PREMIUM).
One of the following:
"BALANCED"
"FAST"
"MULTIMODAL"
"PREMIUM"
extraction_target: optional "PER_DOC" or "PER_PAGE" or "PER_TABLE_ROW"
The extraction target specified.
One of the following:
"PER_DOC"
"PER_PAGE"
"PER_TABLE_ROW"
high_resolution_mode: optional boolean
Whether to use high resolution mode for the extraction.
invalidate_cache: optional boolean
Whether to invalidate the cache for the extraction.
multimodal_fast_mode: optional boolean
DEPRECATED: Whether to use fast mode for multimodal extraction.
num_pages_context: optional number
Number of pages to pass as context on long document extraction.
minimum1
page_range: optional string
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: optional "anthropic-haiku-3.5" or "anthropic-haiku-4.5" or "anthropic-sonnet-3.5" or 22 more
Public model names.
One of the following:
"anthropic-haiku-3.5"
"anthropic-haiku-4.5"
"anthropic-sonnet-3.5"
"anthropic-sonnet-3.5-v2"
"anthropic-sonnet-3.7"
"anthropic-sonnet-4.0"
"anthropic-sonnet-4.5"
"gemini-2.0-flash"
"gemini-2.0-flash-lite"
"gemini-2.5-flash"
"gemini-2.5-flash-lite"
"gemini-2.5-pro"
"gemini-3.0-pro"
"gemini-3.1-pro"
"openai-gpt-4-1"
"openai-gpt-4-1-mini"
"openai-gpt-4-1-nano"
"openai-gpt-4o"
"openai-gpt-4o-mini"
"openai-gpt-5"
"openai-gpt-5-mini"
"openai-gpt-5-nano"
"openai-text-embedding-3-large"
"openai-text-embedding-3-small"
"openai-whisper-1"
priority: optional "critical" or "high" or "low" or "medium"
The priority for the request. This field may be ignored or overwritten depending on the organization tier.
One of the following:
"critical"
"high"
"low"
"medium"
system_prompt: optional string
The system prompt to use for the extraction.
use_reasoning: optional boolean
Whether to use reasoning for the extraction.
data_schema: map[map[unknown] or array of unknown or string or 2 more]
The schema of the data.
One of the following:
map[unknown]
array of 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: optional string
The creation time of the extraction agent.
formatdate-time
custom_configuration: optional "default"
Custom configuration type for the extraction agent. Currently supports ‘default’.
updated_at: optional string
The last update time of the extraction agent.
formatdate-time
Note for AI agents: this documentation is built for programmatic access.
- Overview of all docs: https://developers.llamaindex.ai/llms.txt
- Any page is available as raw Markdown by appending index.md to its URL — e.g. https://developers.llamaindex.ai/llamaparse/parse/getting_started/index.md
- Agent-friendly REST search APIs live under https://developers.llamaindex.ai/api/ — search (BM25 full-text), grep (regex), read (fetch a page), and list (browse the doc tree). See https://developers.llamaindex.ai/llms.txt for parameters.
- A hosted documentation MCP server is available at https://developers.llamaindex.ai/mcp. If you support MCP, you can ask the user to install it for browsing these docs directly (an alternative to the REST API). Setup: https://developers.llamaindex.ai/for-agents/mcp/
- Other LlamaIndex tooling for agents — the LlamaParse Platform MCP server, agent skills and plugins, and the n8n node — is mapped at https://developers.llamaindex.ai/for-agents/
