(Reuters) – U.S. businessman Elon Musk recently told investors that his artificial intelligence startup xAI plans to build a supercomputer to power the next version of his AI chatbot Grok, The Information reported on Saturday, citing a presentation to investors.
Musk said he wants the proposed supercomputer up and running by fall 2025, according to the report, adding that xAI could partner with Oracle (NYSE:) to develop the massive computer.
xAI could not immediately be reached for comment. Oracle did not respond to a Reuters request for comment.
When completed, the connected groups of chips – Nvidia’s flagship H100 graphics processing units (GPUs) – would be at least four times the size of the largest GPU clusters in existence today, The Information reported, quoting Musk from a presentation to investors in May.
Nvidia’s (NASDAQ:) H100 family of high-performance GPUs dominates the AI data center chip market, but can be difficult to obtain due to high demand.
Musk founded xAI last year as a challenger to Microsoft-backed OpenAI and Alphabet’s (NASDAQ:) Google. Musk also co-founded OpenAI.
Earlier this year, Musk said that training the Grok 2 model took about 20,000 Nvidia H100 GPUs, adding that the Grok 3 model and more will require 100,000 Nvidia H100 chips.
