HPE Cray XD670
A 5U server carrying eight NVIDIA H200 SXM GPUs at 700 W each with 141 GB of HBM3e apiece, driven by a single two-socket node of 5th generation Intel Xeon Scalable processors at up to 400 W, with 32 DDR5 DIMMs across eight channels per socket at up to 5,600 MHz. It is air cooled with a direct liquid cooling option. HPE has taken six first-place results with it in MLPerf Inference v5.1, including computer vision and LLM chat and text generation. Two things this row deliberately does not claim: HPE describes the GPUs only as "8x NVIDIA H200 SXM" and never names the HGX baseboard, though eight H200 SXM at 700 W is precisely the HGX H200 8-GPU configuration, and the compute figures shown are NVIDIA's for that baseboard rather than any HPE claim; and HPE publishes component TDPs but no system power figure, so total power is left blank rather than summing 8 x 700 W and 2 x 400 W, which would count silicon only and understate the machine. Part numbers are also not recorded, since none appears on HPE's page.
We will point you at suppliers who have it. Free, and no signup.
Systems built on this design
11 configurationsBuilt on the NVIDIA HGX H200 baseboard, so accelerator performance matches the reference. Chassis, CPU, memory and cooling are this vendor's own.
System Details
GPU Configuration
System Specifications
Precision Performance Breakdown
| Precision | System Performance | Per GPU | Efficiency |
|---|---|---|---|
| 272 TFLOPS | 34 TFLOPS | — | |
| 536 TFLOPS | 67 TFLOPS | — | |
| 3,956 TFLOPS | 495 TFLOPS | — | |
| 7,916 TFLOPS | 990 TFLOPS | — | |
| 7,916 TFLOPS | 990 TFLOPS | — | |
| 15,832 TFLOPS | 1,979 TFLOPS | — | |
| 15,832 TOPS | 1,979 TOPS | — |
Where a vendor states a basis, the figure shown is the dense one, and any figure it publishes is listed separately. Vendors commonly headline the sparse number instead: for NVIDIA's 2:4 structured sparsity that is exactly twice the dense figure, though other vendors' sparse modes do not all follow that ratio.
Powered by NVIDIA H200
This system utilizes 8 × NVIDIA H200 SXM 141GB GPUs, each delivering exceptional performance for AI and HPC workloads.
Per GPU TDP
700W
Per GPU Memory
141 GB
Process Node
4nm
Architecture
Hopper
Frequently asked questions
How much does the Cray XD670 cost?
HPE sells the Cray XD670 on quote through OEMs and resellers, and nobody rents a whole one by the hour. Renting its 8 H200 accelerators separately at the cheapest live rate Flopper tracks, $2.09 per GPU-hour across 21 providers, comes to about $16.72 an hour for the equivalent capacity. H200 rental prices →
How much power does the Cray XD670 require?
HPE has not published a power figure for the Cray XD670, and Flopper does not derive one from the accelerator count because the rest of the system draws power too.
How many GPUs are in the Cray XD670?
8. The Cray XD670 carries 8 NVIDIA H200 SXM 141GB accelerators in a rack form factor, and every throughput figure on this page is the total across them. H200 specifications →
Documentation & Resources
Need this system?
Tell us the quantity, timeframe and region. We reply by email with who can supply it.
Flopper Spec Sheet
HPE Cray XD670 specifications, generated from our database. Printable.
Vendor product page
HPE Cray XD670 documentation
HPE Compute XD690 QuickSpecs (a50009220enw, V6)
HPE • 2026-07-06
Typical Use Cases
The HPE Cray XD670 runs 8× H200 GPUs over NVIDIA NVLink with NVSwitch across the eight SXM modules, delivering 15,832 TFLOPS FP8 dense.