Isometric view of Amsterdam
🧡Open AI Research from the Netherlands

Open LLM Research Made in NL

A Netherlands-based open research collective publishing models, datasets, and findings on large language models — any language, any domain, fully open source.

Why openoranje?

World-class LLM research, fully open, rooted in the Dutch spirit of openness and pragmatism.

Rigorous Research
We publish models and papers with full methodology, training details, and reproducible results.
Any Model, Any Domain
From multilingual translation to domain-specific fine-tunes — we explore wherever the research leads.
Truly Open
Models, datasets, and training code released publicly. Science that anyone can build on.

Latest Releases

Our most recent models and research publications.

Modelv1.0
OpenOranje/TweeTaal-nl-en-0.6B
Our flagship 0.6B parameter model, fine-tuned for Dutch to English and English to Dutch translation.
0.6B params•November 2024

Join the Open Research Movement

Contribute to our research, collaborate on models, or follow our work. Together we're advancing open AI from the Netherlands for the world.