Custom.MT

Overview

Custom.MT is an AI translation platform that supercharges your localization workflows with the leading MT and GenAI models. Manage your stock and custom translation models and connect them to linguistic tools — evaluate and compare MT and LLM engines, and streamline your entire translation operation.

Evaluate and compare DeepL, Google, GPT-4, custom, and other models using the Best Engine tool in the Custom.MT Console. See BLEU and WER scores, example segments, and engine rankings to pick the right model for your localization workflow. Tag processing routines help you preserve tags in software strings during AI translation or clean them out.

Custom.MT works with your translation and content systems through plugins for Trados, Smartling, memoQ, Shopware, XTM Cloud, and Smartcat, plus a powerful API. Train to improve model accuracy with terminology and style, measure model performance by language, domain, and linguist with automated and human evaluations using BLEU, WER, and COMET metrics, and deploy MT on your hardware for confidential content that must not flow into the cloud.

Beyond the platform, Custom.MT offers GenAI services: machine translation model fine-tuning, MT evaluation, on-premise MT, translation memory (TMX) cleaning, automatic post-editing that cuts linguistic audit costs by up to 70%, data acquisition, open-source MT model building, and language data anonymization.

Core capabilities

Best Engine tool

Evaluate and compare DeepL, Google, GPT-4, and custom models with BLEU, WER, and COMET

Model management

Manage stock and custom translation models in one console

Tag processing

Preserve or clean tags in software strings during AI translation

Integrations

Plugins for Trados, Smartling, memoQ, Shopware, XTM Cloud, and Smartcat plus API

On-premise MT

Deploy MT on your hardware for confidential content

Automatic post-editing

Cut linguistic audit costs by up to 70%

Model fine-tuning

Improve accuracy with terminology and style training

Data services

Language dataset acquisition, TMX cleaning, and data anonymization

Use cases

Machine translation management

MT engine evaluation

Localization workflows

On-premise MT deployment

Translation model fine-tuning