Renaissance Labs

Atlas / Europe

Academic lab · Oslo, Norway

Language Technology Group, University of Oslo

University of Oslo research group in natural language processing behind the NorBERT models, the NorBench benchmark and Norwegian generative models.

Norwegianlanguage modelsbenchmarksdatasets

Oslo, Norway

Type
Academic lab
Based in
Oslo, Norway
Key papers
3
Featured projects
8

About

The Language Technology Group (LTG) is part of the Department of Informatics at the University of Oslo, led by Lilja Øvrelid. Its members include Stephan Oepen, Erik Velldal, Andrey Kutuzov, Yves Scherrer, Pierre Lison and David Samuel. The group works on natural language processing and large language models, including for Norwegian.

It released NorBERT in January 2021 and NorBERT 2 in February 2022, then the NorBERT 3 family and the NorBench benchmark in 2023. The group has also trained generative Norwegian models such as NorMistral and NorBLOOM. Projects it participates in include HPLT, OpenEuroLLM, NorwAI, Integreat and MediaFutures.

The group coordinates the Nordic Language Processing Laboratory and publishes code and models on GitHub and Hugging Face.

Mission and ambitions

LTG develops language technology and language models, with a focus on Norwegian and openly available resources for evaluation and training.

Known for

  1. 01

    NorBERT language models

  2. 02

    NorBench benchmark

  3. 03

    NorMistral and NorBLOOM models

Milestones

  1. 2021

    NorBERT

    Released the first large-scale Norwegian contextualised language model, described at NoDaLiDa 2021.

  2. 2022

    NorBERT 2

    Released an updated model trained on about 15 billion words of C4 and NCC.

  3. 2023

    NorBERT 3 and NorBench

    Released a family of four model sizes together with a benchmark for Norwegian.

  4. 2023

    NorMistral and NorBLOOM

    Trained generative Norwegian language models.

Open source

GitHubgithub.com/ltgoslo
  • Norwegian Named Entities annotations on top of NDT (Norwegian Dependency Treebank) (72 stars)

  • Official implementation of "GPT or BERT: why not both?" (64 stars)

  • Norwegian Review Corpus (51 stars)

  • Simple library to work with pre-trained ELMo models in TensorFlow (51 stars)

Hugging Facehuggingface.co/ltg

Research papers

Notable releases

Openness

Models, corpora and code are public on Hugging Face and GitHub under mostly open licences.

People

Listed from the lab's own pages and public profiles.

Contact and community

Contact through the group website.

Sources

Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.

Labs nearby