Elon Musk expands Memphis AI cluster to 1.1M GPUs by 2026

Editorial illustration: An aerial cutaway of a data center shows rows of dark server cabinets extending into a translucent wireframe expansion, with a river, bridges and distant skyline behind it.

In brief

  • Musk targets 1.1 million H100-equivalent accelerators at Memphis Colossus 2 cluster by end of 2026
  • Google signed $920M/month contract for 110,000 GPUs from October 2026 through June 2029
  • Facility powered by temporary gas turbines and Tesla Megapacks; permanent 1.2 GW gas plant under construction

Scale and Infrastructure

Elon Musk announced plans to more than double the Nvidia chip count at the Colossus 2 AI computing cluster by the end of 2026. The expansion would push the cluster toward more than 1.1 million H100-equivalent accelerators, cementing it as one of the largest AI infrastructure buildouts in North America.

The facility already had reached roughly 1 GW of AI compute capacity by mid-2026. The full Colossus 2 buildout is modeled for an IT load of around 946 MW, requiring substantial power infrastructure to sustain operations at scale.

The cluster now integrates multiple sites, including the original Colossus 1 facility alongside Colossus 2 and newer additions like Minihard and Macrohardrr. This multi-site approach allows for distributed capacity while maintaining centralized operational oversight.

Power and Energy

The facility is primarily powered by temporary gas turbines and Tesla Megapacks, a temporary solution that addresses immediate energy demands. A permanent 1.2 GW gas-fired power plant is under construction in Southaven to replace this temporary generation equipment.

Regulators have ordered the phase-out of 69 temporary turbines by mid-2027, creating a hard deadline for the permanent plant to come online. This transition underscores the regulatory scrutiny surrounding large-scale AI infrastructure buildouts and their energy footprint.

Major Commitments

Google signed a contract in June 2026 for approximately 110,000 GPUs, paying a rate of $920 million per month from October 2026 through June 2029. The deal represents one of the largest GPU procurement commitments from a major cloud provider to a single facility.

Anthropic has also secured capacity at the facility, signaling that the cluster is attracting interest beyond Google. These commitments validate the expansion strategy and provide revenue certainty for ongoing buildout costs.