deepset-ai/FARM
A Python framework for fast transfer learning with pretrained language models like BERT and RoBERTa, focused on industrial NLP applications.

FARM (Framework for Adapting Representation Models) provides a streamlined way to fine-tune pretrained language models for downstream NLP tasks. It builds on Hugging Face transformers and PyTorch, offering prebuilt components for common tasks including question answering, named entity recognition, and text classification. The framework targets enterprise use cases with a focus on simplicity and production readiness, though active development has moved to the Haystack project.
Frequently asked
- What is deepset-ai/FARM?
- A Python framework for fast transfer learning with pretrained language models like BERT and RoBERTa, focused on industrial NLP applications.
- Is FARM open source?
- Yes — deepset-ai/FARM is open source, released under the Apache-2.0 license.
- What language is FARM written in?
- deepset-ai/FARM is primarily written in Python.
- How popular is FARM?
- deepset-ai/FARM has 1.8k stars on GitHub.
- Where can I find FARM?
- deepset-ai/FARM is on GitHub at https://github.com/deepset-ai/FARM.