List Table Extractors API Definition
Tabular data extraction plays a crucial role in document processing, enabling efficient indexing and retrieval of structured information. The List Table Extractors API allows users to query available table extractors, ensuring seamless integration with Vectara's document processing pipeline. This API helps users identify the extractors supported by the platform, along with their specifications, to optimize document indexing.
Table extractors process documents to extract and structure tabular content. Users can choose from multiple extractors, each providing different extraction capabilities, or rely on the default extractor provided by Vectara.
List Table Extractors Request and Response Details
To retrieve the available table extractors, send a GET
request to
/v2/table_extractors
. This request returns a list of all supported table
extractors, their names, descriptions, and whether they are the default
extractor for document indexing.
Example response
{
"table_extractors": [
{
"name": "gmft",
"is_default": false,
"description": "GMFT table extraction service",
"generation": {
"llm_name": "gpt-4o",
"prompt_template": "[\n{\n\"role\": \"system\",\n\"content\": \"You are a helpful assistant tasked with summarizing tables...",
"model_parameters": {
"temperature": 0,
"max_tokens": 1024,
"top_p": 1,
"frequency_penalty": 0,
"presence_penalty": 0
}
}
},
{
"name": "textract",
"is_default": true,
"description": "AWS Textract service",
"generation": {
"llm_name": "gpt-4o",
"prompt_template": "[\n{\n\"role\": \"system\",\n\"content\": \"You are a helpful assistant tasked with summarizing tables.. .",
"model_parameters": {
"temperature": 0,
"max_tokens": 1024,
"top_p": 1,
"frequency_penalty": 0,
"presence_penalty": 0
}
}
}
]
}
Error responses
- 403 Forbidden – The user does not have permission to list table extractors.
- 500 Internal Server Error – A server-side error occurred while retrieving extractors.
REST 2.0 URL
List Table Extractors Endpoint
Vectara exposes an HTTP endpoint to retrieve the supported table extractors:
https://api.vectara.io/v2/table_extractors