tencent cloud

知识引擎原子能力

产品简介
产品概述
应用场景
购买指南
计费概述
购买方式
欠费说明
续费说明
退费说明
快速入门
操作指南
原子能力 API 调用
API 文档
History
Introduction
API Category
Making API Requests
Atomic Capability APIs
Data Types
Error Codes
常见问题
相关协议
服务等级协议
服务特别条款
隐私政策
数据处理和安全协议
开源许可声明
联系我们

Data Types

聚焦模式
字号
最后更新时间: 2025-12-24 20:14:53

CreateReconstructDocumentFlowConfig

Configuration information for creating intelligent document parsing tasks.

Used by actions: CreateReconstructDocumentFlow.

Name Type Required Description
TableResultType String No The form in which tables in a Markdown file are returned.
- 0: returned as MD.
- 1: returned as HTML.
The default is 0.
ResultType String No The format of the returned results of intelligent document parsing.
- 0: only returns full-text MD.
- 1: only returns the OCR original Json for each page.
- 2: only returns the MD of each page.
- 3: returns the full-text MD and the original OCR Json of each page.
- 4: returns full-text MD and MD of each page.
The default value is 3 (returns the full-text MD and the original OCR Json of each page).

CreateSplitDocumentFlowConfig

Configuration information for creating intelligent document splitting task.

Used by actions: CreateSplitDocumentFlow.

Name Type Required Description
EnableMllm Boolean No Whether to enable mllm.
MaxChunkSize Integer No Max segment length.

DocumentUsage

The usage of document splitting task.

Used by actions: GetSplitDocumentResult.

Name Type Description
PageNumber Integer Page number of the document splitting task.
TotalTokens Integer Total number of tokens consumed by the document splitting task.

EmbeddingObject

Vector.

Used by actions: GetEmbedding.

Name Type Required Description
Embedding Array of Float No Vector.

Message

Session content.

Used by actions: QueryRewrite.

Name Type Required Description
Role String No Role.
Content String No Content.
ReasoningContent String No Chain of thought content. The ReasoningConent parameter only supports output parameters, and is only returned by the deepseek-r1 model.

ReconstructDocumentFailedPage

Document parsing failure record.

Used by actions: GetReconstructDocumentResult, ReconstructDocumentSSE.

Name Type Description
PageNumber Integer Page number that failed to parse.

ReconstructDocumentSSEConfig

ReconstructDocumentSSE feature configuration parameters.

Used by actions: ReconstructDocumentSSE.

Name Type Required Description
TableResultType String No The form in which tables in a Markdown file are returned.
- 0: returned as MD.
- 1: returned as HTML.
The default is 0.
MarkdownImageResponseType String No The form in which images in a Markdown file are returned.
- 0: returned as URL.
- 1: only return the text content extracted from the image in markdown.
The default value is 0.
ReturnPageFormat Boolean No Whether the Markdown file contains page number information.
PageFormat String No The custom output page number style. {{p}} is a placeholder for the page number. Enable ReturnPageFormat to take effect. If empty, the default style is: page {{p}}.

Usage

Consumption.

Used by actions: GetEmbedding, QueryRewrite, RunRerank.

Name Type Description
TotalPages Integer Number of document pages.
InputTokens Integer Number of input tokens.
OutputTokens Integer Number of output tokens.
TotalTokens Integer Total number of tokens.

帮助和支持

本页内容是否解决了您的问题?

填写满意度调查问卷,共创更好文档体验。

文档反馈