Image Information Extraction

Guide API Reference

Pricing Releases

Upload an image and a prompt, and get the structured extraction result.

POST/v1/ocr/image/extract

Authorization

AuthorizationstringheaderRequired

HTTP: Bearer Auth

Security Scheme Type: http
HTTP Authorization Scheme: Bearer API_key，Used to verify account information, can be viewed in Project Management > API Key .

Request Header

Content-Typeenum<string>Default:multipart/form-dataRequired

Request body is form data. Usemultipart/form-datato upload fields and file.

Available options:multipart/form-data

Request Bodymultipart/form-data

model string Required

Model ID. Current available value:u1-ocr-extract

sceneenum<string>

Specify different scenes. Values and use cases are as follows (default: general).

general：General scene (invoice/receipt/reimbursement/cost list extraction and reconciliation).
medical：Medical scene (medical records/test reports/prescriptions/cost statements/charge lists and detailed expense document-card verification).

image file Required

Image file to be extracted.

Supported formats: JPG, JPEG, PNG, BMP
Size limit: ≤ 10MB

prompt string Required

Prompt used to guide the model to extract specific information.

Example:Please extract the information from the image using the following schema: {"invoice_no":"Invoice Number","date":"Invoice Date","items":[{"name":"Item Name","total":"Amount"}]}

Response Body Structure

contentstring

Text content returned by the model.

usageobject

Token usage statistics.

usage.prompt_tokensinteger

Input token count

usage.completion_tokensinteger

Output token count

usage.total_tokensinteger

Total token count

base_respobject

Status code and details of this request

base_resp.status_codeinteger

Status codes:

0, Success
100001, Missing required
261001, Invalid image format. Only JPG, JPEG, PNG, BMP are supported
261002, Image size exceeds 10MB limit
261003, Image corrupted or cannot be parsed
261101, Upstream model timeout
261999, Internal system error

base_resp.status_msgstring

Status message. "success" when status_code=0, otherwise the error reason.

OCR Extract

curl --request POST \
  --url https://maas-api.hivoice.cn/v1/ocr/image/extract \
  --header 'Authorization: Bearer <api_key>' \
  --header 'content-type: multipart/form-data' \
  --form model=u1-ocr-extract \
  --form scene=general \
  --form 'image=@[object Object]' \
  --form 'prompt=Please extract the information from the image using the following schema: {"invoice_no":"Invoice Number","date":"Invoice Date","items":[{"name":"Item Name","total":"Amount"}]}'

200

{
	"content": "{"invoice_no":"25117000001021449288","date":"2025年08月07日","items":[{"name":"*运输服务*客运服务费","total":"105.95"}]}",
	"usage": {
		"prompt_tokens": 547,
		"completion_tokens": 61,
		"total_tokens": 608
	},
	"base_resp": {
		"status_code": 0,
		"status_msg": "success"
	}
}