guillaume-be/rust-bert
Rust-native NLP pipelines exposing transformer models (BERT, GPT-2, etc.) for tasks including question answering, translation, and text generation.
Velocity · 7d
+1.3
★ / day
Trend
→steady
star history
This repository provides Rust implementations of transformer-based language models including BERT, DistilBERT, GPT-2, RoBERTa, ELECTRA and BART. It exposes both model base architectures and task-specific heads for common NLP pipelines such as question answering, named entity recognition, translation, summarization, and text generation. The library uses tch-rs or ONNX runtime bindings for backend inference with support for GPU acceleration.