API documentation-Document Mind(Document Mind)-阿里云帮助中心

The OpenAPI for this product (Document Mind / 2022-12-29) uses the V2 RPC-style request body and signing mechanism. For more information about signing, see V2 RPC-style request body and signing mechanism. We provide SDKs for common programming languages. You can download an SDK to call the API without needing to handle the signing process yourself. If the available SDKs do not meet your needs, you can integrate the API using the self-signing method. This process is complex and takes approximately five working days to complete. For expert guidance during the integration, join the service DingTalk group at 1137 000 1915. Before you call the API, you must prepare your Alibaba Cloud account and an AccessKey pair. These credentials are required to access the API using client tools, such as the SDK or the command-line interface (CLI). For more information, see Create an AccessKey pair.

Note

To call the APIs for this product, ensure that the account associated with your AccessKey pair has the permission to access Document Mind. In the RAM console, click Add Permissions, search for the AliyunDocmindFullAccess permission, and then add it. This action grants all AccessKey pairs that belong to the RAM account the permission to access Document Mind. For more information, see Service authentication guide.

If an Alibaba Cloud account does not have this permission, the AccessKey pairs that belong to the account cannot be used to access Document Mind.

Document understanding

These APIs provide various features for document parsing and information extraction.

API	API overview
Document parsing (LLM-based)	Extracts layout, text, and other content from documents.
Digital document parsing	Parses information from the semi-structured content of a document.
Intelligent document parsing	Extracts the hierarchical structure, text, key-value (KV) fields, and style information from documents.
Intelligent table parsing	Extracts table styles, content, text KVs, and table KVs.
Document extraction	Automatically extracts key information from documents and tables. Returns the information as generic KV structured content.

Document format conversion

API	API overview
Convert PDF to Word	Converts a PDF file to an editable Word document. It accurately recognizes text and preserves the original layout and style.
Convert PDF to image	Converts each page of a PDF file to an image. It processes PDF documents that contain up to 100 pages.
Convert PDF to Excel	Converts a PDF file to an editable Excel document. It accurately recognizes text and preserves the original layout and style.
Convert image to Excel	Converts images to an editable Excel document. It supports up to 30 images per conversion. It accurately recognizes text and preserves the original layout and style.
Convert image to Word	Converts images to an editable Word document. It supports up to 30 images per conversion. It accurately recognizes text and preserves the original layout and style.
Convert image to PDF	Merges up to 30 images into a single PDF file.