Companies/xAI
USA
xAI
AI lab that built the Colossus training cluster in Memphis, Tennessee. Ownership has reportedly changed since the cluster was built; this dataset records the operator as xAI and does not assert a corporate parent.
Systems this company appears in
- Systems
- 1
- Aggregate Rmax
- 0.0 GFlop/s
- Aggregate power
- 0 kW
- First appearance
- 2025-02
- Most recent
- 2025-02
- Concentration
- -
systems it appears in
where disclosed
largest is xAI Atlanta
Aggregate Rmax is the sum of the systems this company supplies any part of. It is not revenue and not attributable share: an integrator and a memory vendor in the same machine both count its full FLOPS here. Attribution weighted by role is Phase 4.
Role breakdown
| Role | Systems | Rmax in those systems |
|---|---|---|
| Integrator | 1 | 0.0 GFlop/s |
Appearances
| System | Supplied as | Shipped under | Rmax | Rank |
|---|---|---|---|---|
| xAI Atlanta USA · 2025-02 | Integrator | xAI | - | unranked |
AI datacenters operated
| System | Status | Accelerators | Stated power (MW) |
|---|---|---|---|
| xAI Colossus | operational | 100,000 | 150 |
| xAI Colossus 2 | operational | 220,000 | 1,276 |
| xAI Atlanta | operational | 12,448 | 17 |
Figures come from the AI datacenter metrics; blank means none recorded.
Compute deals
| Date | Role | Counterparty | Type | MW | Value ($bn) | Source |
|---|---|---|---|---|---|---|
| 2026-05 | Seller | Anthropic | Cloud contract | - | - | sec.gov |
| 2026-05 | Buyer | Valor Equity Partners | Other | - | 20.206 | sec.gov |
| 2025-11-19 | Buyer | HUMAIN | Other | 500 | - | middleeastainews.comw.media |
Headline values are as announced and are not additive. See all compute deals.
Sites operated
| Site | Location | Cooling | Commissioned |
|---|---|---|---|
| xAI Colossus (Memphis) | Memphis, TN, USA | - | 2024 |
| xAI Colossus 2 (Whitehaven, Memphis) | Memphis, TN, USA | - | 2026 |
| QTS Atlanta Metro Campus (xAI) | Atlanta, GA, USA | - | 2025 |
Related news
- xAI's Colossus 2 reaches 200MW of cooling capacity in six months · SemiAnalysisThe Memphis and Southaven build reached the milestone in about six months against roughly fifteen months for comparable Oracle, Crusoe and OpenAI projects, and is targeted at over 1GW of GB200 NVL72 capacity. It is a data point on how much faster a single well capitalized operator can now stand up gigawatt class capacity than the shared infrastructure neoclouds this dataset also tracks.