HPE logo

HPE ProLiant Compute XD685 (HGX B300)

rack
8× B300
NVLink 5.0 / NVSwitch 4.0 across the HGX B300 baseboard; 4x FHHL PCIe Gen5 x16 slots for fabric, NIC or DPU
2025
FP64
10
TFLOPS
FP32
600
TFLOPS
Power
—
kW Total
Memory
2100
GB Total

The eight-way XD685 in its NVIDIA HGX B300 configuration, which requires the high-power liquid-cooled chassis S6Y07A rather than the standard DLC one. This is the denser sibling of the air-cooled XD690, which reaches the same eight Blackwell Ultra GPUs in 10U where this does it in 5U. Two fifth-generation AMD EPYC processors, 24 DIMM slots and up to 6 TB of DDR5-6400, eight EDSFF NVMe E3.S drives. HPE publishes no system power draw for the XD685, only twelve 3,000 W Titanium supplies with ten deployable, so unlike the XD690 no figure is recorded; the compute shown is NVIDIA's specification for the HGX B300 baseboard rather than an HPE claim.

We will point you at suppliers who have it. Free, and no signup.

FP64
10
TFLOPS
FP32
600
TFLOPS
TF32
9,000
TFLOPS
FP16
18,000
TFLOPS

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: Blackwell Ultra
Interconnect: NVLink 5.0 / NVSwitch 4.0 across the HGX B300 baseboard; 4x FHHL PCIe Gen5 x16 slots for fabric, NIC or DPU

System Specifications

Form Factor: rack
Total Power: —
Total Memory: 2.1 TB
Memory Bandwidth: 64 TB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
10 TFLOPS 1.3 TFLOPS —
600 TFLOPS 75 TFLOPS —
9,000 TFLOPS 1,125 TFLOPS —
18,000 TFLOPS 2,250 TFLOPS —
18,000 TFLOPS 2,250 TFLOPS —
36,000 TFLOPS 4,500 TFLOPS —
108,000 TFLOPS 13,500 TFLOPS —
FP6 36,000 TFLOPS 4,500 TFLOPS —

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 B300

This system utilizes 8 × NVIDIA B300 SXM 262GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

1400W

Per GPU Memory

263 GB

Process Node

4nm

Architecture

Blackwell Ultra

Documentation & Resources

Flopper Spec Sheet

HPE ProLiant Compute XD685 (HGX B300) specifications, generated from our database. Printable.

Vendor product page

HPE ProLiant Compute XD685 (HGX B300) documentation

View ↗

HPE Compute XD690 QuickSpecs (a50009220enw, V6)

HPE • 2026-07-06

View Document ↗

Typical Use Cases

AI/ML Training
High-Performance Computing
Data Analytics

The HPE ProLiant Compute XD685 (HGX B300) runs 8× B300 GPUs over NVLink 5.0 / NVSwitch 4.0 across the HGX B300 baseboard; 4x FHHL PCIe Gen5 x16 slots for fabric, NIC or DPU, delivering 36,000 TFLOPS FP8 dense.

© 2026 Flopper.io - Compare the hardware powering AI