What we’ve noticed:
In conversations with IT leaders, project managers, and business departments, it becomes clear that it’s not just about how automated document processing works in theory – but about very practical implementation questions.
That’s why in this first part, we focus on three key aspects that decision-makers are currently most interested in:
What AI models do we use?
Is processing fully cloud-based or also available on-premises?
Who is the cloud provider?
Three questions. Three clear answers. Let’s go.
What AI models does ExB use – rule-based or LLMs?
Short & sweet:
At ExB, we rely on Large Language Models (LLMs) – not rule-based systems or proprietary in-house solutions.
In detail:
LLMs like GPT or PaLM have made tremendous progress in recent years. They’re pre-trained on vast amounts of text, bring broad world knowledge, and can be flexibly adapted to specific requirements – for example, using sample documents from your domain.
Compared to rule-based systems or custom-trained models, LLMs offer significantly greater generalization capabilities – a clear advantage in the complex world of logistics documents. They don’t just recognize what they’ve been explicitly taught – they also “understand” variations, new phrasings, and complex contexts.
Our philosophy: We are technology-neutral and choose the best model for each use case. Whether the focus is on maximum accuracy or optimal cost-efficiency – we pick what fits best. New, more powerful models are being developed every month. We remain flexible and decide on a case-by-case basis.
Is processing fully cloud-based – or also available on-premises?
Our platform is cloud-based – for good reason. LLMs require massive computing power. In the cloud, this power can be scaled as needed: high load? More resources kick in. Low usage? Resources scale down automatically. This significantly reduces costs and ensures a low-maintenance, capital-light operation.
At this time, we do not offer on-premises solutions due to the high complexity: typically 3 to 5 times more effort than a pure cloud solution. For small and medium-sized businesses, this is usually not economically viable.
Therefore, our standard offering is cloud-based processing – GDPR-compliant, fast, and efficient.
Which cloud provider is used – and where does the data go?
Our infrastructure typically runs on AWS – with servers located in Frankfurt, Germany.
This ensures that all data stays within Europe – a crucial factor for many of our customers.
That said, we are open: depending on the use case, we also leverage services such as Google’s Vertex AI. The choice depends on the model in use and its specific requirements.
What this means for you: Your data remains in secure, certified European data centers. And you benefit from state-of-the-art infrastructure – with zero setup effort on your end.
Conclusion: Why this matters to you
The difference between traditional document processing and modern, AI-powered IDP solutions lies not just in the technology – but in the real-world impact:
With ExB, you get ready-to-use, highly accurate LLMs that work without complex rule definitions or time-consuming training.
Our cloud-based infrastructure ensures everything runs flexibly, securely, and at scale – without hidden complexity and without tying up your internal IT resources.
And best of all: You don’t need to be an AI expert to unlock the full potential. We handle the tech – you focus on your processes.
Want to see how ExB can work in your company?
Then talk to us. In a short demo, we’ll show you how easy it is to get started.
👉 Book an appointment now
