📚 API Documentation

The Receipt Scan AI API enables developers to extract structured data from receipts and invoices.

Get started by adding your payment information for a free trial for the first month and
generate your API Key from the Dashboard Page.

You will need an Open AI or Gemini API key to use this API.

Every request must include your apiKey (generated in your Receipt Scan AI dashboard) and a modelKey from your chosen AI provider (OpenAI or Gemini).

Important: Keep your keys secret. Rotate them immediately if compromised.

📌 How to Get API Keys

• Gemini API (Google AI Studio): Get your Gemini API key
• OpenAI API: Get your OpenAI API key

Use your Receipt Scan AI apiKey together with either an OpenAI or Gemini modelKey.

Extract structured data (items, total, tax, date) from a receipt or invoice image.

Example Request:

POST https://extractinvoicedatahttp-loshywauza-uc.a.run.app

{
  "apiKey": "your-api-key",
  "modelKey": "openai-or-gemini-model-key",
  "model_type": "openai", // or "gemini"
  "image": "base64-encoded-image"
}
          

Example Response:

{
  "items": [
    "VEGETABLE & FRUITS",
    "WATER BOTTEL 5LTR",
    "AL MARAI 200ML"
  ],
  "invoice_date": "2025-09-07",
  "total_price": 14,
  "tax": 1.83,
  "isImageAReceipt": true
}
          

You can generate or rotate API keys from the Dashboard Page

Strict Mode: when enabled, malformed AI responses will be rejected with an error. Disable it for fallback parsing.
For example, if you send an image that is not a receipt image or the API failed to detect any product from that image, the API will respond with a 400 error code if strict mode is enabled.

Common error codes:

• 400 → Missing required fields / invalid model type
• 403 → Subscription inactive or expired
• 500 → Internal server error

Example Error:

{
  "error": "Inactive or missing subscription."
}
          

• After adding your payment information, you get a 1-month Free Trial
• You can extract data from an unlimited number of receipt images
• After the trial, you will be charged 9.99$ per month

JavaScript (Node.js)

async function extractInvoice(base64Image, apiKey, modelKey, modelType) {
  const response = await fetch(
    "https://extractinvoicedatahttp-loshywauza-uc.a.run.app",
    {
      method: "POST",
      headers: { "Content-Type": "application/json" },
      body: JSON.stringify({
        apiKey,
        modelKey,
        model_type: modelType,
        image: base64Image
      })
    }
  );

  const data = await response.json();
  console.log("Extracted:", data);
}
    

🐍 Python

import requests
import json

def extract_invoice(base64_image, api_key, model_key, model_type):
    url = "https://extractinvoicedatahttp-loshywauza-uc.a.run.app"
    payload = {
        "apiKey": api_key,
        "modelKey": model_key,
        "model_type": model_type,
        "image": base64_image
    }
    response = requests.post(url, headers={"Content-Type": "application/json"}, data=json.dumps(payload))
    print("Extracted:", response.json())
    

🐹 Golang

package main

import (
  "bytes"
  "encoding/json"
  "fmt"
  "net/http"
  "io/ioutil"
)

func extractInvoice(base64Image, apiKey, modelKey, modelType string) {
  url := "https://extractinvoicedatahttp-loshywauza-uc.a.run.app"

  payload := map[string]string{
    "apiKey": apiKey,
    "modelKey": modelKey,
    "model_type": modelType,
    "image": base64Image,
  }

  body, _ := json.Marshal(payload)
  resp, _ := http.Post(url, "application/json", bytes.NewBuffer(body))
  defer resp.Body.Close()
  
  data, _ := ioutil.ReadAll(resp.Body)
  fmt.Println("Extracted:", string(data))
}
    

⚛️ ReactJS (Frontend)

import { useState } from "react";

export default function InvoiceUploader() {
  const [output, setOutput] = useState("");

  async function handleExtract(base64Image, apiKey, modelKey, modelType) {
    const res = await fetch(
      "https://extractinvoicedatahttp-loshywauza-uc.a.run.app",
      {
        method: "POST",
        headers: { "Content-Type": "application/json" },
        body: JSON.stringify({
          apiKey,
          modelKey,
          model_type: modelType,
          image: base64Image,
        }),
      }
    );

    const data = await res.json();
    setOutput(JSON.stringify(data, null, 2));
  }

  return (
    <div>
      <button onClick={() => handleExtract("BASE64_IMAGE", "YOUR_API_KEY", "MODEL_KEY", "openai")}>
        Extract Invoice
      </button>
      <pre>{output}</pre>
    </div>
  );
}
    

💻 cURL

To call the API you must pass your apiKey, the model key (either OpenAI or Gemini), and the image encoded as a Base64 string.

Convert your image to Base64 first:

• Mac/Linux:
  base64 invoice.png | tr -d '\n' > invoice.b64.txt
  (The Base64 string will be written into invoice.b64.txt.)
• Windows PowerShell:
  $imgPath = "C:\path\to\invoice.png"
$base64 = [Convert]::ToBase64String((Get-Content $imgPath -Encoding Byte))
$base64 | Out-File "C:\path\to\invoice.b64.txt"

Example request with cURL:

curl -X POST "https://extractinvoicedatahttp-loshywauza-uc.a.run.app" \
  -H "Content-Type: application/json" \
  -d '{
    "apiKey": "YOUR_API_KEY",
    "modelKey": "OPENAI_OR_GEMINI_MODEL_KEY",
    "model_type": "openai",
    "image": "BASE64_ENCODED_IMAGE"
  }'

💻 PowerShell (Invoke-RestMethod)

$imgPath = "C:\path\to\invoice.png"
$base64  = [Convert]::ToBase64String((Get-Content $imgPath -Encoding Byte))

$body = @{
  apiKey     = "YOUR_API_KEY"
  modelKey   = "OPENAI_OR_GEMINI_MODEL_KEY"
  model_type = "openai"
  image      = $base64
} | ConvertTo-Json -Depth 3

$response = Invoke-RestMethod -Uri "https://extractinvoicedatahttp-loshywauza-uc.a.run.app" `
  -Method Post `
  -Headers @{ "Content-Type" = "application/json" } `
  -Body $body

$response

This PowerShell example reads the image, encodes it to Base64, and sends it to the API with Invoke-RestMethod.