Yandex metrika counter
Nvidia developing new chip to boost AI speed
Source: Reuters

Nvidia is preparing to launch a new processor aimed at helping OpenAI and other clients build faster and more efficient artificial intelligence systems, according to sources familiar with the matter.

The chipmaker is developing a new platform focused on “inference” computing — the type of processing that enables AI models to generate responses to user queries, News.Az reports, citing the Wall Street Journal.

The system is expected to be introduced at Nvidia’s upcoming GTC developer conference in San Jose next month.

The report said the platform will include a chip designed by startup Groq.

Earlier this month, Reuters reported that OpenAI has expressed dissatisfaction with the speed at which Nvidia’s hardware generates responses for certain tasks, including software development and AI-to-software communication. According to one source, OpenAI is seeking new hardware that could eventually cover around 10% of its inference computing requirements.

The ChatGPT developer has also held discussions with chip startups including Cerebras and Groq about supplying processors for faster inference, sources said. However, Nvidia reportedly secured a $20 billion licensing agreement with Groq that effectively ended OpenAI’s talks with the startup.

In September, Nvidia announced plans to invest up to $100 billion in OpenAI as part of a broader deal that granted the chipmaker a stake in the company while providing OpenAI with capital to purchase advanced chips.


News.Az 

By Nijat Babayev

Similar news

Archive

Prev Next
Su Mo Tu We Th Fr Sa
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31