R

Robbert 2023 Dutch Large

Developed by DTAI-KULeuven
RobBERT-2023 is a Dutch language model based on the RoBERTa architecture, developed by KU Leuven, Ghent University, and TU Berlin, and is one of the state-of-the-art language models for Dutch.
Downloads 627
Release Time : 12/5/2023

Model Overview

RobBERT-2023 is the 2023 version of the Dutch RobBERT model, trained on the OSCAR 2023 dataset, available in base and large versions, and excels in Dutch natural language processing tasks.

Model Features

Latest Data Training
Trained on 2022 data, including new vocabulary such as COVID-19 and the latest world knowledge
Performance Improvement
Significant improvement over previous models and BERTje in the DUMB benchmark
Dual Version Option
Offers base and large versions (355 million parameters) to meet different needs

Model Capabilities

Text Understanding
Text Generation
Text Classification
Named Entity Recognition
Question Answering

Use Cases

Academic Research
Dutch NLP Research
Used for Dutch natural language processing related research
Excellent performance in the DUMB benchmark
Commercial Applications
Dutch Customer Service System
Building Dutch intelligent customer service and question answering systems
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase