Our goal

MMT will overcome four technology barriers that still hinder the wide adoption of currently available MT software by end-users and language service providers:

  • MMT will be a ready to install application that will not require any initial training phase.
  • The MMT system will manage context automatically so that it will not require building domain-specific systems.
  • MMT will enable scalability of data and users so that no more expensive ad-hoc hardware installations are needed.
  • MMT will create a data collection infrastructure that accelerates the process of filling the data gap between large web companies and the machine translation industry.

Download on Github

MMT release is available on GitHub!
Download it now and try ModernMT, you can read the installation instructions here.

Download on Github v.2.0.1 - MMT release for Ubuntu 14.04 and 16.04 Watch ModernMT on Github

The best features from generic and custom systems

Generic MT engines
  • Commercial, closed technology
  • Good quality for generic translations
  • Highly scalable
  • Based on huge amounts of data
  • Ready to use
Custom MT engines
  • Open source technology
  • High quality for specific domains
  • Not scalable
  • Comparatively small data
  • Long set-up time

Data Collection

Collecting billions of words from MyMemory, TAUS Data and Common Crawl

Big Data Architecture

Updating Moses to support highly scalable infrastructure and big data processing

Open Source

Technology based on open source components and released as open source

What Makes MMT unique

Real time
adaptation to content

No more training
or set-up time

Designed to scale
for data and users

Stay up-to-date with MMT

Your name (required)

Your email (required)