CS

Cerebras CS-3 Inference Cluster (8-node)

cluster
8× WSE-3
On-wafer fabric per engine, 214 Pb/s; cluster networking via dedicated NET cabinets
2024
Power
240.0
kW Total
Memory
GB Total

An eight-node Cerebras inference cluster, where a node is one CS-3 system. Cerebras provisions it as five cabinets: four WSE cabinets holding two CS-3 systems each, plus one NET cabinet. Provisioned heat load is approximately 240 kW, of which about 220 kW is carried by water and 20 kW by air. Note that this is Cerebras's own provisioned figure and is higher than eight times the 27 kW system power, because it includes the networking cabinet. The cluster holds 352 GB of on-chip SRAM and no DRAM. Cerebras rates each engine at 125 petaFLOPS but footnotes the figure as sparse, and because the sparsity is unstructured no dense equivalent is derived.

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

System Details

GPU Configuration

GPU Model: Cerebras WSE-3
GPU Count: 8 GPUs
Architecture: Wafer Scale Engine 3
Interconnect: On-wafer fabric per engine, 214 Pb/s; cluster networking via dedicated NET cabinets

System Specifications

Form Factor: cluster
Total Power: 240.0 kW
Total Memory:
Memory Bandwidth: 168000000 GB/s

Powered by Cerebras WSE-3

This system utilizes 8 × Cerebras WSE-3 GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

Per GPU Memory

Process Node

TSMC 5nm

Architecture

Wafer Scale Engine 3

Documentation & Resources

Official Datasheet

Cerebras CS-3 Inference Cluster (8-node) technical specifications

Download PDF

Cerebras CS-3 datasheet: system, rack and inference cluster specifications

Cerebras • Latest version

View Document ↗

Typical Use Cases

AI/ML Training
High-Performance Computing
Data Analytics

The Cerebras CS-3 Inference Cluster (8-node) runs 8× WSE-3 GPUs over On-wafer fabric per engine, 214 Pb/s; cluster networking via dedicated NET cabinets.