Deepl publishes new NVIDIA chips for the entire internet translation in 18 days

Depp said on Wednesday that she was publishing one of the latest Nafidia Systems that would allow German operation to start translating the entire Internet in just 18 days.

This decreased sharply from 194 days before.

Deepl is an emerging company that has developed its artificial intelligence models It competes with Google translation.

At the same time, NVIDIA is looking to expand the customer base in its chips – which is designed to run artificial intelligence applications – beyond Passover such as Microsoft and Amazon.

It also highlights how startups use high -end NVIDIA products to build artificial intelligence applications, which are seen as the next step after constituent models, such as those designed by Openai.

Cologne -based company publishes the NVIDIA system known as DGX SuperPod. Both DGX SuperPod server contains 36 B200 Grace Blackwell Superchips, one of the company’s latest products on the market. NVIDIA chips are required to train and operate huge AI models, such as those designed by Deepl.

“The idea is, of course, providing a lot of mathematical power to our research scientists to build more advanced models,” Stefan Misskin, the chief scientist in Deeb, told CNBC.

Mesken said that the upgraded infrastructure will help boost current products such as the clarification, which the company launched this year. Clarify the tool that asks user questions to ensure that the context is integrated into the translation.

“It was not technically possible until recently with the developments that we have made in our upcoming efforts. This is now possible. So these are the types of developments that we continue to search for,” said Misskin.