Feb 27 (Reuters) – Nvidia plans to launch a new processor aimed at helping OpenAI and other customers build faster and more efficient artificial intelligence systems, the Wall Street Journal reported on Friday, citing people familiar with the matter.
Nvidia is developing a new system for “inference” computing, a form of processing that allows AI models to respond to queries, the report said.
The report quoted people familiar with the matter as saying that the new platform will be unveiled at the Nvidia GTC developer conference in San Jose next month and will use chips designed by the startup Groq.
Reuters could not immediately verify the report. Nvidia and OpenAI did not immediately respond to Reuters’ requests for comment.
Reuters reported earlier this month that OpenAI was dissatisfied with the speed with which Nvidia’s hardware provided ChatGPT users with answers to certain types of questions, such as software development and artificial intelligence communicating with other software.
One of the sources told Reuters that it needs new hardware that could eventually meet about 10% of OpenAI’s future inference computing needs.
ChatGPT makers have discussed working with startups such as Cerebras and Groq to provide chips for faster inference, two sources said. But a source told Reuters that Nvidia had ended OpenAI talks by striking a $20 billion licensing deal with Groq.
In September, Nvidia said it intended to invest $100 billion in OpenAI as part of a deal that would give the chipmaker a stake in the startup and provide OpenAI with the cash it needs to buy advanced chips.
(Reporting by Mihika Sharma in Bengaluru; Editing by Tom Hogue)
