CLUSTERCluster · GB200 platform · partial site profile
- Operator / site
- OpenAI / Oracle Cloud Infrastructure
- Country
- United States
- Profile scope
- Abilene site · partial public profile
- Data checked
- 14 September 2026
Stargate · Abilene
OpenAI / Oracle Cloud Infrastructure
CPU Accelerator Interconnect
i
Reading this map
Stargate · Abilene node architecture
Stargate · Abilene is operated by OpenAI / Oracle Cloud Infrastructure in United States. This profile covers Abilene site · partial public profile. Specifications are listed below; anything the operator has not published is itemised in the documented-fields list.
GB200 platform · partial site profile
- CPU model
- NVIDIA Grace
- Accelerator model
- NVIDIA Blackwell (GB200)
- CPU ↔ accelerator link
- NVLink-C2C
- Form factor
- NVIDIA GB200 NVL72 rack — 36 Grace CPUs and 72 Blackwell GPUs per rack, direct-to-chip liquid cooled
Scope and uncertainty
The site is the first of the Stargate programme, on the Lancium Clean Campus in Abilene, Texas. Crusoe, the site developer, publishes campus figures rather than a machine configuration: eight buildings, about 4 million square feet and 1.2 GW total capacity, with direct-to-chip liquid cooling. NVIDIA documents the GB200 NVL72 rack as 36 Grace CPUs and 72 Blackwell GPUs in one NVLink domain. Neither party publishes a rack count, system count or GPU total for Abilene, so none is recorded.
Documented fields: 4 / 23
Every field this catalog checks. Fields listed above have a value; the remainder are shown here. A field marked absent is not publicly available in the sources we could verify, which does not mean the hardware is missing. Fields that cannot apply, such as benchmark results for an unranked system, are excluded from the total.
Cluster — 0 / 3
- Manufacturer
- Ranked core count
- Theoretical peak (PFlop/s)
- Measured power (kW)
- Energy efficiency (GFlop/s per watt)
- Operating system
- Network topology
GB200 platform · partial site profile — 4 / 20
- System / node model
- Compute nodes of this type
- CPU model
- CPU sockets per node
- Cores per socket
- Physical CPU cores
- CPU note
- Accelerator model
- Physical accelerators per node
- Host / unified memory
- Accelerator memory
- Accelerator ↔ accelerator link
- On-node fabric
- CPU ↔ accelerator link
- Network interfaces per node
- Interface model
- Interface speed
- Interface-to-accelerator topology
- Aggregate scale-out per node
- GPUDirect RDMA
- Management / storage networking
- Local storage
- NVMe data disks
- Max remote data disks
- Uncached disk IOPS
- Uncached disk throughput (MBps)
- Form factor