apache/opennlp

Provides machine learning-based NLP tasks (tokenization, POS tagging, named entity recognition, parsing, coreference resolution) using classifiers like Maximum Entropy, Perceptron, and SVM, with modular Java components. Integrates with Apache Flink, Spark, and NiFi for distributed pipelines, plus UIMA annotators and ONNX model support with optional GPU acceleration. Offers both programmatic Java API and CLI interfaces with pre-trained multilingual models.

1,586 stars. Actively maintained with 33 commits in the last 30 days.