Technical overview of the ModelFront API

Read our API documentation or read more about risk prediction and its use cases.

If you are a current client with an urgent issue or question, email sup[email protected] now.

Getting started

To get started with ModelFront, create an account to get an API key and access to the sandbox. Then you're ready to play with it and hit our REST API.


The risk API provides real-time risk predictions on each machine translation. You get a risk prediction score between 0.0 and 1.0 on each machine translated segment, typically a sentence or a phrase.

You set your own quality threshold to control the balance of quality, scale and speed for each type of text.

Read our API documentation

Error types

ModelFront is built to catch many kinds of errors, both the common ones and the rare but painful ones, including:

Named entities

Errors translating names of persons, places, organisation and products


Revered meaning due to missing or inserted negation


Offensive or ambiguous translations


Untranslated words or phrases


Garbage input or output characters, words or phrases


Many other types of translation errors, for example translating from the wrong language


ModelFront is built to support any language pair with a sizable open corpus:

Afrikaans, Albanian, Arabic, Armenian, Azerbaijani (Latin and Perso-Arabic), Basque, Belarusian, Bengali, Bosnian, Bulgarian, Burmese, Catalan, Cebuano, Chechen, Chinese (Simplified and Traditional), Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Kyrgyz, Korean, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Marathi, Mongolian, Nepali, Norwegian (Bokmal and Nynorsk), Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Scots, Serbian, Serbo-Croatian, Slovak, Slovenian, Spanish, Sundanese, Swahili, Swedish, Tagalog, Tajik, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, Yoruba


You can use our generic base model or get your own models trained on your own monolingual data, parallel data, post-edited data or labelled data.

Under the Partner plan, we can also invest in custom development like crawling and human labelling to hit your accuracy goals.

Pricing and plans

ModelFront offers simple volume-based pricing: $200 per 1M source-side characters.

See our plans for more details and options

Confidentiality and security

Client and user data privacy are very important to us.

The text in your API requests is automatically deleted by default. Your data for your custom models are never used for any other models. You can permanently delete your custom models at any time.

We are happy to opt in to a standard one-way non-disclosure agreement (NDA) before you send us your data or discuss your business needs with us.

We additionally offer on-premise deployment under our Partner plan. To discuss private clouds and on-premise deployment, please contact us.