NbAiLab/borealis-open-270m
NbAiLab/borealis-open-270m is a 270-million parameter instruction-tuned language model from the Borealis family, developed by the National Library of Norway (Nasjonalbiblioteket, NB). Based on Google's Gemma 3 architecture, this model is specifically fine-tuned for Norwegian-centric instruction following, including tasks like drafting, summarization, and question answering. It is designed for use cases involving Norwegian language processing and assessment, offering a compact solution for localized NLP applications.
Loading preview...
Borealis Open 270M Overview
NbAiLab/borealis-open-270m is a 270-million parameter instruction-tuned model from the Borealis family, developed by the National Library of Norway (Nasjonalbiblioteket, NB). This model is an "open release" variant, meaning it does not include material from specific rights-holder agreements, unlike its full Borealis counterparts. It is built upon the Gemma 3 architecture, specifically fine-tuned from google/gemma-3-270m-it.
Key Capabilities
- Norwegian-centric Instruction Following: Optimized for tasks such as drafting, summarization, and question answering in Norwegian.
- Language Assessment: Useful for evaluating and improving Norwegian writing style and quality.
- Safety Alignment: Incorporates "prompt baking" to distill system prompt behavior into model weights, balancing usefulness with safer outputs.
- Compact Size: At 270M parameters, it offers a lightweight solution for Norwegian NLP tasks.
Good For
- Norwegian Language Applications: Ideal for assistant-style tasks, content generation, and text analysis in Norwegian, Bokmål, and Nynorsk.
- Research and Development: Suitable for evaluating language coverage and behavior specific to Norwegian dialects.
- Resource-Constrained Environments: Its smaller size makes it practical for deployment where computational resources are limited.