Xaynia: New energy- and cost-efficient search AI for business

Press Release

May 10

Written By Clara Herdeanu

European AI company Xayn presents search and recommendations AI Xaynia

Xaynia helps companies to be more energy- and cost-efficient through Small Language Models

The suite of Natural Language Processing models combines accuracy, personalization, and privacy
Semantic Search understands search queries and context similar to humans

BERLIN, 10 May 2023 – The research-based AI company Xayn presents today Xaynia, the enterprise semantic search, and recommendation AI. It combines accuracy, energy- and cost-efficiency with privacy. Business clients across different industries can easily implement the AI plug-in via API to provide their customers with a smooth and individually personalized user experience. Xaynia moves with its Natural Language Processing (NLP) technology from matching to understanding its users’ search queries and their context and delivers highly accurate results tailored to individual preferences.

Energy- and cost-efficiency through Small Language Models

“While Large Language Models like ChatGPT have created quite a buzz they also consume huge amounts of energy – at the cost of the environment. This is not only a problem for the climate, but it also increases the costs for business clients immensely, especially when used at scale. Additionally, we don’t know what happens with all of the personal data," says Leif-Nissen Lundbæk, PhD (CEO & Co-Founder of Xayn). "Our suite of multilingual NLP models Xaynia solves these problems for search and recommendations by moving from Large to Small Language Models that are even so efficient that they can run on a mobile phone instead of the latest supercomputer.”

Keeping the accuracy of Large Language Models, Xayn compressed its AI to maximize the energy efficiency of Xaynia. This way, Xaynia can process Big Data such as millions of documents in large databases while minimizing energy consumption and costs at the same time. In a recently published paper, Xayn tested its own model extensively on energy efficiency comparing it to other popular transformer models which are used to understand users’ search intentions and indexed contents. In these settings, Xaynia was up to 20 times more energy efficient.

Xayn is validating its technology with different media companies such as the German public-broadcast TV station ZDF. The media institution is aiming to improve its media library ZDFmediathek by personalizing the recommendations for its users in a more privacy-protecting and energy-efficient way.

Diverse applications across different industries

With personalization proven to increase revenues and achieve higher retention rates, Xayn’s versatile semantic search and recommendations plug-in addresses the needs of companies with large outward-facing knowledge bases from sectors such as media, e-commerce, or travel. Catering to the users who don't actively state their preferences, personalized search results, and recommendations ensure content relevance, enhancing the overall user experience. Xaynia can be applied to websites, apps, push notifications, and newsletters, creating a cohesive and engaging experience for users across various platforms.

European standards in transparency, privacy, and sustainability

The Berlin-based company originated in Oxford and is committed to the European values of transparency, privacy, and sustainability and is pushing for Europe’s technological sovereignty. Therefore, it is developing Xaynia as a privacy-protecting open-source solution. On the paid plan, Xayn’s business clients will benefit from an easy-to-integrate API solution. It enables them to utilize Xaynia and Xayn’s highly efficient infrastructure without the need for machine learning expertise or maintaining a complex content discovery system themselves.

For more information go to www.xayn.com

_________________________

About Xayn

Xayn is pioneering next genAI for large organizations. The European startup builds secure, sovereign, and sustainable AI systems that leverage the power of language models at scale. Xayn develops its Retrieval Augmented Generative AI as energy efficient and privacy-protecting AI solution “Made In Germany”.

Born out of research at Oxford University and Imperial College London, the Berlin-based company is committed to European values of transparency, privacy, and sustainability and develops its AI system as open-source technology. Founded in 2017 by Dr Leif-Nissen Lundbæk, Professor Michael Huth and Felix Hahmann, Xayn’s academic vision remains, with a workforce comprised of more than 35% PhDs. The start-up has received investment funding of 19.5 million EURO from Global Brain Corporation, KDDI Open Innovation Fund, Earlybird, and Dominik Schiener.

Downloads