REST API · Available now

Add text recognition to anything you build

Process handwritten and printed documents programmatically. Submit an image, get structured text back. Any language, any framework — just HTTP.

EU-hosted (Austria) · GDPR-compliant · 50% credit discount · Organisation plan

recognize.py
import requests

resp = requests.post(
    "https://transkribus.eu/processing/v2/processes",
    headers={"Authorization": f"Bearer {TOKEN}"},
    json={
        "config": {"modelId": 38230},
        "image": {
            "imageUrl": "https://your-archive.org/scan-001.jpg"
        }
    }
)
job = resp.json()

What will you build?

The metagrapho API fits into any architecture. Here's what teams ship with it.

Digitise at scale

Batch-process thousands of scans unattended. Submit images, poll for completion, download structured output. Run overnight, wake up to searchable text.

Async job queue — submit and forget
URL or base64 image input
PAGE XML, ALTO, and plain text output
batch-submit.sh
# Submit 500 images in parallel
for img in $(cat image-urls.txt); do
  curl -s -X POST \
    'https://transkribus.eu/processing/v2/processes' \
    -H "Authorization: Bearer $TOKEN" \
    -H "Content-Type: application/json" \
    -d '{"config":{"modelId": 38230},
         "image":{"imageUrl": "'$img'"}}' &
done
wait
echo "All jobs submitted."

Documentation

Everything you need to integrate the metagrapho API.

Authentication

The metagrapho API uses the OpenID Connect protocol for authentication. Obtain an access token, then include it as a Bearer token in all API requests.

OpenID Connect configuration: https://account.readcoop.eu/auth/realms/readcoop/.well-known/openid-configuration

EndpointsOpen Swagger UI →

Good to know

API credits cost 50% less than Transkribus UI credits. Pay-per-use billing is available upon request.

To use a private model, your API account must be added to the collection where the model is linked.

Accepted formats: JPEG, TIFF, PNG. Maximum file size: 20 MB per image.

The Transkribus Advantage

Train in the UI. Deploy via API.

Your domain experts create ground truth and train custom AI models in Transkribus — no code required. Your developers deploy those models at scale through the API. One platform, two workflows.
Domain experts create ground truth in a visual editor
One-click model training — no ML expertise needed
Deploy trained models instantly via API
300+ community models also available

Coming Summer 2026

The documentation above covers the current metagrapho API. We're building something bigger.

Developer Preview

The Transkribus Developer Platform

A complete developer portal with multiple APIs, interactive documentation, and OpenAPI specs. The current metagrapho API stays backwards-compatible — the new platform gives you more power, more endpoints, and a modern developer experience.
New Job API with webhooks and richer config
File API for uploads, exports, and collections
Sites API for publishing digital editions
Full OpenAPI specs and interactive docs
new-job-api.py
# New Job API (coming soon)
import transkribus

job = transkribus.jobs.create(
    type="htr",
    input={"fileId": "file_abc123"},
    config={
        "modelId": 38230,
        "outputFormats": ["page-xml", "text"]
    },
    webhook="https://your-app.com/callback"
)
50% discount
vs Transkribus UI credits
Scalable
Thousands of pages per day
300+
Public AI models
EU-hosted
GDPR-compliant, Austria

Ready to integrate text recognition?

Available with Organisation plans or upon request. All processing on our servers in Austria, EU — fully GDPR-compliant.