# deepset-ai/FARM

:house_with_garden: Fast & easy transfer learning for NLP. Harvesting language models for the industry. Focus on Question Answering.

Repository: https://github.com/deepset-ai/FARM
Canonical: https://ross.abutalabs.com/products/deepset-ai-farm
Homepage: https://farm.deepset.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: language-models, bert, nlp, deep-learning, transfer-learning, pytorch, nlp-library, nlp-framework, xlnet-pytorch, ner, question-answering, pretrained-models, roberta, germanbert
Archived: true
Last push: 2023-12-20T21:18:02+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2604, "days_push": 987, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived, no_readme
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1752, forks 244 (observed 2026-08-28T04:05:31.772082+00:00)

## What it is
FARM is a Python framework for fine-tuning and evaluating transformer-based language models for NLP tasks, with a focus on question answering. It provides modular building blocks for transfer learning with models like BERT, RoBERTa, and XLNet in industry settings.

## Use cases
- fine-tune BERT for question answering
- train a named entity recognition model with transformers
- evaluate pretrained language models on NLP tasks
- build a document QA system over my own data
- transfer learning for German NLP models
- fine-tune RoBERTa for text classification

## When to choose
- you need fine-grained control over fine-tuning transformer models for QA or NER
- you want a PyTorch-based framework with experiment tracking and evaluation built in
- you are working with German or multilingual BERT-style models

## When to avoid
- you want a production-ready QA pipeline rather than a training framework (prefer deepset's Haystack)
- you need the latest LLM features or active community support
- you only need inference with pretrained models without fine-tuning

## Facets
- artifact type: framework
- maturity: maintenance
- function: machine-learning, nlp, llm-training, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: transfer-learning, transformers, question-answering, named-entity-recognition, bert, pytorch, finetuning, natural-language-processing

## Member repositories
- deepset-ai/FARM (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:31.772082+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:28:27.053235+00:00, confidence not recorded.
  - registry_pypi: https://pypi.org/pypi/farm/json (fetched 2026-08-29T11:06:20.633360+00:00, sha e76c0117c300)
- Data as of 2026-08-30T08:39:29.467469+00:00.