Download Area

Home > Transformer Models

CTranslate2 (free) Download Full | **UPDATE

Fast inference engine for Transformer models - CTranslate2

CTranslate2 (free) Download Full | **UPDATE

Published Date: 2024-04-14

CTranslate2 Free Download

CTranslate2 is a free and open-source statistical machine translation system developed by Carnegie Mellon University. It is a powerful tool that can be used to translate text between over 100 different languages. CTranslate2 is based on the Moses machine translation system, and it uses a variety of techniques to improve translation quality, such as phrase-based translation, hierarchical phrase-based translation, and factored translation models. CTranslate2 is available as a command-line tool, a web service, and an API.

CTranslate2 is a popular choice for machine translation because it is accurate, efficient, and easy to use. It is also free and open-source, which makes it a great option for developers and researchers. If you are looking for a powerful and reliable machine translation system, CTranslate2 is a great choice.


CTranslate2: CTranslate2 is a C++ and Python library for efficient inference with Transformer models. The project implements a custom runtime that applies many performance optimization techniques such as weights quantization, layers fusion, batch reordering, etc., to accelerate and reduce the memory usage of Transformer models on CPU and GPU. The execution is significantly faster and requires less resources than general-purpose deep learning frameworks on supported models and tasks thanks to many advanced optimizations: layer fusion, padding removal, batch reordering, in-place operations, caching mechanism, etc. The model serialization and computation support weights with reduced precision: 16-bit floating points (FP16), 16-bit integers (INT16), and 8-bit integers (INT8). The project supports x86-64 and AArch64/ARM64 processors and integrates multiple backends that are optimized for these platforms: Intel MKL, oneDNN, OpenBLAS, Ruy, and Apple Accelerate.