The announcement comes from companies that helped build the xAI facility. Greater Memphis Chamber It was held during a lunch organized by. Colossus was deployed in the summer of this year with 100,000 Nvidia H100 GPUs in an incredible four months. At that time, it was announced that this was the largest artificial intelligence supercomputer. When scaled with 1 million GPUs, it will surpass existing supercomputers by a large margin. El Capitan, currently the most powerful supercomputer, has approximately 43 thousand AMD CPUs and 43 thousand GPUs.
xAI will soon increase Colossus’ capacity to 200,000 GPUs and next summer as of 300.000 GPUs aims to achieve. Most of the GPUs are expected to be supplied from Nvidia. xAI, “it isHe did not explain when 1 million GPUs would be reached. Elon Musk, “I’m just trying to complete 1% of a pathetic Kardashev Type I civilization!” he said. Early next year for Musk’s xAI supercomputer to $1.08 billion will cost Nvidia Blackwell GB200 It was also brought to the fore that he met with Huang and offered a higher price than normal in order to get priority access to his servers. But 1 million GPUs costs tens of billions of dollars It will happen.
xAI recently raised $6 billion in new funding, with some of the funding going to Colossus. While the company develops artificial intelligence models called Grok, it offers the chatbot of the same name to paid X users.
Meanwhile, other companies like Meta, OpenAI, and Microsoft have also purchased large quantities of Nvidia’s latest GPUs. In January, Meta CEO Mark Zuckerberg said his company would have “approximately 600,000 H100-equivalent computes” by the end of this year. Meanwhile, the Greater Memphis Chamber stated that Nvidia, Dell and Supermicro Computer, which helped found Colossus, plan to operate in the city.
This news our mobile application Download using
You can read it whenever you want (even offline):