Musk Targets 1.1M GPUs in Memphis as AWS Adds 2M Nvidia Chips

AI Daily Briefing

The episode examines the massive scale of upcoming AI infrastructure expansions, specifically focusing on Elon Musk's xAI cluster and AWS's

Key takeaways

• Elon Musk plans to expand the Colossus II cluster in Memphis to over 1.1 million H100-equivalent GPUs by late 2026.
• Google's GPU contract is estimated at $920 million per month for approximately 110,000 GPUs through June 2029.

Main topics

• xAI Memphis cluster expansion
• AWS and NVIDIA GPU scaling

Notable quotes

Whoever controls the inference stack holds the leash, and right now that's XAI.

Conclusion

The episode highlights a massive-scale arms race in compute capacity s

Transcript preview

Speaker 1 (0:00) Two giants spent the past day counting GPUs by the million. Speaker 1 (0:05) Meanwhile, a drone over New Jersey counted something a lot less flattering. Speaker 3 (0:09) And I've got one question for all of it. How much is actually plugged in? It's AI Daily Briefing, Friday edition. We start with Musk in Memphis. Speaker 1 (0:21) From Diego Almada-Lopez at Crypto Briefing. Speaker 2 (0:24) Elon Musk laid out a concrete timeline for scaling up the Colossus II AI computing cluster in the Memphis, Tennessee and South Haven, Mississippi corridor. committing to more than doubling the facility's NVIDIA chip count before the end of 2026. The expansion would push the cluster toward more than 1.1 million H100-equivalent accelerators. Speaker 3 (0:44) Okay, do the division on the Google deal with me. $920 million a month for roughly 110,000 GPUs is about $8.5 grand per GPU per month. Locked in through June 2029. That's the number I care about. Way more than the million. Speaker 1 (1:08) And the million's a year-end target. Musk says he'll more than double the chip count to over 1.1 million H100 equivalents by December. It's September 25th. Those racks don't exist yet. Speaker 3 (1:21) Right. And look at what's powering the stuff that does exist. Temporary gas turbines and Tesla megapacks. The word temporary is sitting under a 946 megawatt IT load. Speaker 1 (1:34) I'd underline the tenant list, though. Google's paying rent, and Anthropic has capacity there, too. A safety-branded lab training on Musk's power patchwork in Memphis. Whoever controls the inference stack holds the leash, and right now that's XAI. Speaker 3 (1:51) Yeah, no, I'd flip it. At eight grand a GPU, Google's got the leash. Musk needs that check clearing every month to fund the doubling. Speaker 1 (2:01) From Kieran Allen at CloudWars. Speaker 2 (2:03) Now, the two powerhouses have announced a staggering expansion of this existing partnership, which will, among other things, but this is certainly the standout headline, include the deployment of 2 million additional NVIDIA GPUs across AWS infrastructure between 2027 and 2028. The aim, according to Amazon, is to meet surging demand for AI infrastructure from frontier labs, global enterprises, startups, and governments. So stack it Speaker 3 (2:32) on the Memphis number we just did. Colossus 2's 1.1 million, plus 2 million more from AWS.