Libre Translate is an on-prem translation SaaS. It is free and open, with code hosted on github.
You can see our demo at translate.wasyaco.com - but please request an API key if you are planning on using it at any scale.
We can spin up your own instance of Libre Translate, either on your infrastructure or on ours. We install it via docker. Alternatively, you can plug in ssh keys and ip of your own machine, and use wco_hosting_rb service to to manage the install on your own hardware - see the documentation.
LibreTranslate is an open-source machine translation platform designed to provide a free, privacy-focused alternative to proprietary translation services. It supports multiple languages and uses neural network models to perform translation without sending user data to external servers, making it suitable for individuals and organizations that prioritize data security and control. Because it is community-driven and permissively licensed, developers can self-host, customize, and integrate LibreTranslate into applications, workflows, and services without usage limits or subscription fees.
The product offers both a web interface and an API, enabling users to translate text interactively or programmatically. It also includes features such as language detection, batch translation, and optional integration with external model providers. LibreTranslate’s focus on simplicity and accessibility makes it a practical choice for educational projects, research, and businesses seeking an adaptable translation tool that can be deployed on-premises or in private cloud environments.
Most of the time you would use Libre Translate via its API. An API call looks like this:
curl -X POST https://translate.wasyaco.com/translate \
-H "Content-Type: application/json" \
-d '{
"q": "Hello, how are you?",
"source": "en",
"target": "es",
"format": "text"
}'.
.^.