AI-powered
data extraction
for

Build, deploy & scale the most accurate deep learning OCR APIs for any document type, without ML experience.
01
 {
02
   "label": "total_amount",
03
   "value": "169.95",
04
   "confidence": "0.975816"
05
 }
TRANSFORMING AUTOMATION AT

Build accurate data capture APIs, tailored to you.

Invoice OCR

Extract information like invoice number, vendor name, dates, amounts, etc.

Receipt OCR

Extract information like vendor name, date, net amount, receipt number, etc.

ID Card OCR

Extract information like name, date of birth, nationality, ID card number, etc.

Any document

Create a model for extracting information from any document type.

Why choose Cradl?

Any document
Invoices, receipts, ID cards, bank statements. You decide.
Any language
Trainable on any language
regardless of character set.
Best-in-class accuracy
Process documents with human-like precision with powerful ML models.
Real-time processing
Less than one second average processing time.
TECHNOLOGY

Powerful
deep learning
at your fingertips

Fine tune our powerful machine learning models on your own documents to get the most accurate data capture possible.
HOW IT WORKS

AI platform made for engineering teams

Configure model
Define the information you want to extract from your documents. Text fields, numerical fields or enums - you decide.
Add training data
Upload historical data from your database or ERP system, or integrate directly in your existing workflow to collect training data.
Train your model
Select the data you want to train on, and start training your model with the click of a button.
01
02
import pathlib
03
from las import Client
04
05
client = Client()
06
image = pathlib.Path('image.jpeg').read_bytes()
07
document_id = client.create_document(image)['documentId']
08
model_id = 'las:model:e9091228xxxxxxxxxxxxxx'
09
prediction = client.create_prediction(document_id, model_id)
10
Integrate with your solution
Seamlessly integrate your trained model in your own software with easy-to-use SDKs or through our REST API.
CONTINUOUS LEARNING

Always improving with seamless feedback loops

Implement a feedback loop to ensure that your model learns from every document you process, and improve automation and accuracy over time.

Frequently Asked Questions

Who is Cradl for? 
Cradl is created for development teams with advanced data capturing needs. Our users are typically developers in enterprises with high volumes of documents who want to take full advantage of machine learning technology to automate document processes.
What is the accuracy?
A number of factors affect the accuracy of a trained model. In practice, the most important factors are the size and quality of the training dataset.

For documents like invoices and receipts, a well trained model can have field-wise accuracies over 99%. Furthermore, every prediction comes with a corresponding confidence which measures the probability that a prediction is correct. By filtering out low-confidence predictions based on a threshold, you can increase the accuracy by making a trade-off between accuracy and automation.

For example, assume you have a model trained on invoices, and that you want 99.8% accuracy for total amount predictions. Through experimentation, you may find that if you set your confidence threshold to 0.9, then 75% of all total amount predictions have a confidence level above 0.9, and you observe a 99.8% accuracy for these predictions. The remaining 25% can be sent through a manual verification step.
How much data do I need to train a model?
The higher the variation in the documents, the more training data is required. Training a model to process receipts from all over the world requires significantly more training data than if you are training a model to read national ID cards.

We recommend training on at least 5 000 documents. The training data should be as similar as possible to the data you want the model to process. Training data can often be automatically extracted from a database, archive, etc. so that manual document labelling is not required.
Do I need to have experience with machine learning to use Cradl AI?
No. As long as you can use our REST APIs, Cradl AI takes care of the heavy lifting.

Ready to start building?

Build your model for free. Choose a plan when you are ready to launch.