Skip to main content

Enterprise-grade Document Handling Platform

EASY. SCALABLE. MEASURABLE

Max Mustermann
Musterstraße 3, 12345 Musterstadt
Kundennummer: 123456790
More features
More extensive integration
Customised AI

CONTACT US

Customers from various branches trust lector:

client
client
client
client
client

Our AI platform for
fully automated document processing

Enhanced with customised enterprise features for maximum integration and security.

CLASSIFICATION

Customised solutions for document separation

  • Use of classic classification models or pre-trained LLMs with efficient prompt engineering
  • No need for separator sheets or stamps
  • Reliable recognition even from photos

Data extraction

Fully automated extraction of relevant information

  • Precise capture even for complex, unstructured documents, including handwritten values
  • Use of multimodal models for processing text and images​

CLOUD
SERVICE

  • Fast onboarding
  • Access to the latest LLMs
  • Standard import/export APIs
  • Scalable

USAGE-BASED PRICING

TRY IT FOR FREE

HYBRID

ON-PREMISE
INSTALLATION

  • Highest level of data protection
  • Use of existing IT infrastructure
  • Integration of existing models
  • Customised integrations

FIXED PRICING

REQUEST A DEMO

SIMPLE. SCALABLE. MEASURABLE.

REQUEST A DEMO

ADVANCED ENTERPRISE FEATURES

All the features of Lector Studio ,customised to your needs:

CUSTOMISED AI

  • We fine-tune customised AI models to customer data.
  • On-premise LLMs

LARGE LANGUAGE MODELS

General

No training required

Quick setup

Can also understand images

Requires prompt engineering

Cost

CUSTOM MODELS

Tailored to use cases

Highly efficient

Training required

Development effort

LEAN LANGUAGE MODELS

Small and fast

Focus on the essentials

Inexpensive

Training necessary

Finite context

COMPLEX LANGUAGE MODELS

Specialised

Focus on the essentials

Affordable

Training required

Finite context

VISUAL MODELS

Reading complex structures

Handwriting recognition

Recognition of special markings

Training & large models required

Hallucination possible-Finite context

ADVANCED INTEGRATION

  • Modular structure with various import and export options
  • REST API, webhook exporter, email importer​
  • Integration with external OICD providers

EXTENDED MONITORING
& INSIGHTS

  • Transparent and traceable workflows
  • Usage statistics and monitoring of the system and AI components​

ENTERPRISE SUPPORT

  • Premium SLA with fixed response and resolution times
  • Premium support for integration into existing systems

Performance KPIs

90

accuracy in document classification

500

documents processed

80

average time savings in processing

FREE DEMO

Simply request a demo of the platform!

We will show you in a webcast
(Teams, Webex, Zoom etc.),
how processing works onlector.ai.

Our team will be happy to answer all your questions.

REQUEST A DEMO

“

"As a supra-local social welfare provider, the district of Lower Franconia receives a large number of extensive, but also heterogeneous and unstructured incoming mail. Since autumn 2022, we have been working with lector.ai GmbH to process our incoming mail in accordance with TR-RESISCAN, classify it in detail and automatically forward it to the relevant department. We are very satisfied with the current progress of the project and the work of lector.ai, and feel that we are on the right track to optimising our processes with the efficient use of an AI software solution, with highly competent support."

Anja WeigertHead of Department for Service, Digitalization & eGovernment
“

‘Do you still open and scan your letters yourself and want to keep it that way? Then you know that recognising and correctly interpreting what is written is not so easy, especially in freely formulated texts. Here, lector.ai helps us at over 25 health insurance companies to process incoming mail automatically and make processes more efficient.’

Jan KekeritzCDO itsc
“

‘Thanks to lector.ai's customised software solution and in-depth expertise, we have been able to optimise our internal processes by over 50% and save significant time and costs. The collaboration has always been professional, efficient and reliable, so we can wholeheartedly recommend lector.ai and are already planning further joint projects.’

Maren AleffMaren Aleff, Managing Director

FAQ

Why lector?

With lector, the days of manually reading and typing documents are over! Here, you can automate your document processing in just a few simple steps. You also benefit from constant improvements to AI models because we always incorporate the latest ones into our platform.

See for yourself — try it out for free!

What types of documents can lector process?

Our understanding of the term ‘document’ is very broad. We process not only invoices, forms and letters, but also emails, images and tables. We can process the following file types:

  • Documents: PDF, DOC, DOCX, RTF, ODT
  • Tables: XLS, XLSX, ODS, CSV
  • Images: JPEG, PNG, TIFF, HEIC, HEIF
  • Web & Markup: HTML, XML, Markdown
  • Communication: MSG (Outlook), EML (E-Mail), vCard
  • Archives: ZIP
  • Text:TXT, Plain Text

Which interfaces are available? (Input/output)

  • By default, we can import via API and email. We also offer import from data storage systems, apps, and customised solutions.
  • By default, we export via webhook. We also offer export to document management systems, the cloud, databases and customised solutions.

 

Our interfaces are explained here

Which AI does lector use?

Whichever one you want! We always have the latest models available! The following open and closed source models are currently available to you: Gemini, GPT, Claude, internVL and Qwen. If you are interested in closed-source models, we can offer you the option of GDPR-compliant use via European servers (hosted by Vertex and Azure), or you can connect directly with providers in the USA.

Where is my data stored?

Our platform runs on the OVHcloud infrastructure in Europe, which is hosted in Germany. AI processing via Vertex and Azure takes place on European servers. However, the data is only stored here while it is being processed and is then deleted immediately. Everything is, of course, GDPR-compliant. You can also specify the storage period for your documents on our platform.

Is my data used to train AI?

No. The data is only used by the AI models for processing and is deleted immediately afterwards.

What hardware is required for on-premises operation?

To run lector.ai on-premises, you need a system with Ubuntu as the operating system (Docker may already be pre-installed in the image). The minimum requirements are 16 CPU cores, 64 GB of RAM, and 500 GB of hard disk space. To optimally support AI-based document processing, an Nvidia A100 80 GB or H100 80 GB graphics card is required.

TRANSFORM YOUR DOCUMENT WORKFLOWS!

EASY. SCALABLE. MEASURABLE.

CONTACT US