The new Buildsimple Foundation model

AI · Security

The new Buildsimple Foundation model

A consistent evolution for stable document processes

Last updated: 22. September 2025

At Buildsimple, we are continuously developing our AI platform step by step. Our goal is a solution that works reliably, can be easily integrated into existing processes and does not require large volumes of data. Foundation models make an important contribution to this.

What is a Foundation model?

A Foundation model is a pre-trained base model that has developed a fundamental understanding of language and structure through large volumes of texts, documents and layouts. It recognises contextual relationships, typical document structures and patterns that frequently occur in invoices, forms, emails or contracts.
This previously acquired knowledge makes Foundation models more stable and significantly less dependent on large training datasets. This is precisely why they are becoming increasingly prevalent in document processing.

With the new Buildsimple Foundation model, we are creating an improved technical basis for classification, extraction and future separation functions within the platform.

Why Foundation models are crucial now

Document processes are changing. Companies today receive data from a wide variety of sources and with highly variable structures. Traditional machine-learning approaches often reach their limits, especially when only limited training data is available or documents are not structured uniformly.

Foundation models provide a stable basis for this. They are pre-trained on large volumes of texts, layouts and examples, and bring a fundamental understanding that also helps with new, unfamiliar or incomplete documents. For input management, this means greater robustness, fewer manual rules and less training effort.

The new Buildsimple Foundation model at a glance

The new model is an in-house developed base model created through extensive document training. It understands language, layout structures and thematic relationships, and can categorise new content more quickly.

Key goals of the new Buildsimple Foundation model:

  • A better fundamental understanding of complex document structures

  • Less dependence on large training datasets

  • More stable results with varying layouts

  • Less effort required for setup and customisation

  • A unified basis for the platform's core AI functions

This allows the model to work reliably even when documents look different, have different layouts or are available in varying quality.

Concrete benefits for companies

Companies benefit directly from the model's improved capabilities.

Particularly relevant benefits:

  • Significantly less training effort

  • Greater accuracy in classification

  • Stable results despite different layouts or scan qualities

  • Faster customisations and iterations

  • Fewer manual rules and lower maintenance requirements

  • Better results even with small volumes of data

  • Improved processing of heterogeneous document batches

Technical foundations and how it works

The Foundation model combines various modern methods to better understand and reliably process documents. Each method helps meaningfully connect language, layout and content.

Key components of the model:

  • Pre-training with many real documents to recognise typical patterns and structures

  • A shared vector space for text, images and layout, enabling the model to categorise complete pages

  • A modern AI approach that recognises relationships within a document

  • A learning method that reliably identifies differences and similarities between documents

  • Fine-tuning that delivers good results even with only a few examples

The model supports both work in specialist departments and the requirements of IT and architecture.

Applications in the Buildsimple product

The Foundation model forms the technical basis for several components of the Buildsimple platform.

Areas in which the model is or will be used:

  • Classification: the new engine is already built entirely on the Foundation model.

  • Extraction: the next generation of extraction models also uses the Foundation basis.

  • Document separation: a new separation engine is in development and will use the model for context-based boundary detection.

  • Migration of existing models: existing customer solutions can be migrated to the new model on request.

  • Long-term product strategy: the model serves as a central basis for a wide range of AI functions in Buildsimple.

Conclusion

The new Buildsimple Foundation model is a consistent evolution of our AI platform. It makes document processes more stable, accurate and easier to use, while also forming the basis for future improvements. For companies, this means less effort, higher quality and more reliable results in everyday work with documents.

Newsletter

Expert insights straight to your inbox

Get new articles, webinar dates and practical insights straight to your inbox — compact, relevant and genuinely useful for your day-to-day work.

Enjoyed this article?

Feel free to share it!