SAN FRANCISCO, CA – Unbabel is bringing its industry-leading COMET and QE tools to Widn.AI, making enterprise-grade translation quality evaluation accessible to businesses of all sizes via a simple API.
Recognized as the industry gold standard for translation quality, these AI-powered tools achieve state of the art performance in translation quality evaluation. Using this technology businesses can distinguish top tier from low quality translations, cut down review time, boost efficiency, and reduce operational costs.
Up until now most AI translation flows were “blind” when it comes to quality, and companies need to either trust that these are correct or pay for additional human in the loop support. With quality evaluation, this is no longer the case – you can continuously assess the quality of your systems – and only take action if and when required.
What was once the preserve of enterprise companies can now be leveraged by organisations of all sizes through empowering companies to scale their AI powered translation operations with confidence.
Trusted by global enterprises and major technology companies alike, COMET and QE bring different strengths to quality evaluation. COMET’s neural technology matches human judgment across 100+ languages, ensuring translations meet real-world quality standards. QE performs instant quality checks without reference texts—offering real-time insights into translation accuracy.
With Widn.AI’s API, these powerful tools are now within reach for language service providers, developers, and businesses of all sizes—without the complexity.
Vasco Pedro, Co-founder & CEO, Unbabel said: “At Unbabel, we push the limits of AI to deliver world class translation. By integrating our industry-leading COMET and QE tools with Widn.AI, we’re democratizing access to the very best quality evaluation on the market—helping businesses to scale their operations faster and more efficiently than ever before.”
Learn more about how COMET, QE can help your business scale and optimise your translations at Widn.AI
For media enquiries please contact farah.pasha.ext@unbabel.com
ABOUT COMET and QE
COMET is a cutting-edge neural framework for translation quality evaluation, built by Unbabel to match human judgment across 100+ languages. Using advanced AI and large-scale cross-lingual models, it delivers highly accurate predictions of translation quality.
Unlike traditional metrics, COMET learns from human assessments, analyzing both source text and translations to produce quality scores that closely reflect expert evaluations. Combined with Quality Estimation (QE), which enables real-time quality checks without reference translations, COMET sets the bar for translation quality assessment.
ABOUT Widn.Ai
Widn.AI is a powerful, yet straightforward Language AI solution for businesses seeking enterprise-grade translations without enterprise-level costs. Powered by one of the world’s leading multilingual LLMs, Widn.AI delivers natural, authentic translations that capture true meaning, not just words.
Developed by Unbabel’s world-class research team, Widn.AI is built to scale with your business needs, helping organizations scale confidently across languages and markets.
Our goal is simple: help businesses grow faster with world class Language AI.
