Invoice data extraction API: how to extract invoices to JSON

说明与用途

AnyFormat API 将 PDF 发票转换为结构化 JSON，包含标题字段、行项目、置信度得分和来源证据，简化 ERP 集成。

问题

Accounts payable teams need structured invoice data (header fields, line items, confidence scores, evidence) to post to ERP systems, but OCR scripts often return unstructured text or invented values.

结果

The API returns structured JSON with header fields, line items, confidence scores, and evidence, enabling automated ERP integration and reducing manual review.

工作流

• Create a workflow with a parse node and an extract node, supplying a JSON schema for invoice fields.
• Upload an invoice file to the workflow and start a run, receiving a run_id and status.
• Poll the run endpoint until status is "processed" and retrieve the extracted fields from the results JSON.

证据

An invoice data extraction API takes an invoice as a PDF or image and returns structured JSON: header fields such as invoice number, vendor, dates and totals, plus the line items, each value with a confidence score and, in the better ones, the place on the page it was read from.

来源

The API has three steps. Create a workflow with a schema, upload an invoice against it, and fetch the result.

来源