This RoBERTa model was pretrained from scratch on a large Polish corpus using a custom 8k token vocabulary. It serves as a base model for fine-tuning on Polish-specific natural language understanding tasks.
Polish Roberta 8k is an Other AI project. It focuses on providing a strong Polish-language foundation model for downstream NLP tasks such as classification and named entity recognition. It is built as an open-source project for developers. Polish Roberta 8k is open source under the Open Source license. It ships for the web and API.
PKOBP builds and maintains Polish Roberta 8k, and it first shipped in 2025.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do