Extract text from screenshots, scanned images, receipts, labels, and photos.
extract_text
extract_text_with_layout
Receipt parsing
Screenshot indexing
Document text extraction
All ImageHQ processing endpoints are asynchronous. Upon a successful POST, you receive a 202 Acceptedresponse with a job_id. Poll the status endpoint until the state reaches succeeded.
Request Example
import requests
url = "https://api.imagehq.io/v1/ocr"
payload = {
"tool_slug": "image-to-text",
"operation": "extract_text",
"options": {
"language": "eng",
"detect_orientation": True,
"output_format": "txt"
}
}
files = [("files[]", open("image.png", "rb"))]
data = {"request": json.dumps(payload)}
response = requests.post(url, files=files, data=data)
print(response.json())const form = new FormData();
form.append("files[]", file);
form.append("request", JSON.stringify({
"tool_slug": "image-to-text",
"operation": "extract_text",
"options": {
"language": "eng",
"detect_orientation": true,
"output_format": "txt"
}
}));
const response = await fetch("https://api.imagehq.io/v1/ocr", {
method: "POST",
headers: { "Idempotency-Key": crypto.randomUUID() },
body: form
});
const data = await response.json();
console.log(data);const form = new FormData();
form.append("files[]", file);
form.append("request", JSON.stringify({
"tool_slug": "image-to-text",
"operation": "extract_text",
"options": {
"language": "eng",
"detect_orientation": true,
"output_format": "txt"
}
}));
const response = await fetch("https://api.imagehq.io/v1/ocr", {
method: "POST",
headers: { "Idempotency-Key": crypto.randomUUID() },
body: form
});
const data = await response.json();
console.log(data);curl -X POST "https://api.imagehq.io/v1/ocr" \
-H "Idempotency-Key: $(uuidgen)" \
-F "files[]=@image.png" \
-F 'request={"tool_slug":"image-to-text","operation":"extract_text","options":{"language":"eng","detect_orientation":true,"output_format":"txt"}}'$client = new GuzzleHttp\Client();
$response = $client->post("https://api.imagehq.io/v1/ocr", [
"multipart" => [
["name" => "files[]", "contents" => fopen("image.png", "r")],
["name" => "request", "contents" => '{"tool_slug":"image-to-text","operation":"extract_text","options":{"language":"eng","detect_orientation":true,"output_format":"txt"}}']
]
]);require "faraday"
response = Faraday.post("https://api.imagehq.io/v1/ocr") do |req|
req.headers["Idempotency-Key"] = SecureRandom.uuid
req.body = { "files[]" => Faraday::UploadIO.new("image.png", "image/png"), "request" => '{"tool_slug":"image-to-text","operation":"extract_text","options":{"language":"eng","detect_orientation":true,"output_format":"txt"}}' }
endbody := &bytes.Buffer{}
writer := multipart.NewWriter(body)
writer.WriteField("request", `{"tool_slug":"image-to-text","operation":"extract_text","options":{"language":"eng","detect_orientation":true,"output_format":"txt"}}`)
file, _ := writer.CreateFormFile("files[]", "image.png")
_ = file
writer.Close()
http.Post("https://api.imagehq.io/v1/ocr", writer.FormDataContentType(), body)HttpRequest request = HttpRequest.newBuilder()
.uri(URI.create("https://api.imagehq.io/v1/ocr"))
.header("Idempotency-Key", UUID.randomUUID().toString())
.POST(HttpRequest.BodyPublishers.ofString("multipart form data"))
.build();using var form = new MultipartFormDataContent();
form.Add(new StringContent('{"tool_slug":"image-to-text","operation":"extract_text","options":{"language":"eng","detect_orientation":true,"output_format":"txt"}}'), "request");
form.Add(new StreamContent(File.OpenRead("image.png")), "files[]", "image.png");
await httpClient.PostAsync("https://api.imagehq.io/v1/ocr", form);var request = URLRequest(url: URL(string: "https://api.imagehq.io/v1/ocr")!) request.httpMethod = "POST" request.setValue(UUID().uuidString, forHTTPHeaderField: "Idempotency-Key") // Attach multipart files[] and request fields before sending.
{
"queued": {
"id": "job_123",
"status": "queued",
"operation": "ocr",
"tool_slug": "png-to-jpg",
"client_reference_id": "example-123",
"progress": 0,
"current_stage": "queued",
"poll_url": "/v1/jobs/job_123",
"created_at": "2026-05-02T00:00:00Z",
"expires_at": "2026-05-03T00:00:00Z"
},
"completed": {
"id": "job_123",
"status": "succeeded",
"progress": 100,
"inputs": [
{
"filename": "input.png",
"format": "png",
"mime_type": "image/png",
"size_bytes": 420122
}
],
"outputs": [
{
"id": "0",
"filename": "output.jpg",
"format": "jpg",
"mime_type": "image/jpeg",
"size_bytes": 161002
}
],
"warnings": [],
"stages": [
{
"name": "queued",
"status": "succeeded",
"progress": 100
},
{
"name": "processing",
"status": "succeeded",
"progress": 100
}
],
"download_url": "/v1/jobs/job_123/download",
"retention_policy": {
"ttl_hours": 24,
"clamp": true
},
"expires_at": "2026-05-03T00:00:00Z",
"result_json": {
"text": "Extracted text...",
"confidence": 0.91
}
}
}OCR language support depends on configured language packs in your deployment.
Yes. OCR responses include confidence metadata where available.
Yes. OCR can be one step in a pipeline workflow.