Cerebras CS-3 Inference Cluster (32-node)
A thirty-two-node Cerebras inference cluster, provisioned as twenty cabinets: sixteen WSE cabinets holding two CS-3 systems each, plus four NET cabinets. Provisioned heat load is approximately 945 kW, about 875 kW on water and 70 kW on air, Cerebras's own provisioned total including networking. The cluster holds 1.408 TB of on-chip SRAM and no DRAM. Each engine is rated at 125 petaFLOPS, footnoted by Cerebras as sparse; the sparsity is unstructured, so no dense figure is derived.
We will point you at suppliers who have it. Free, and no signup.
System Details
GPU Configuration
System Specifications
Powered by Cerebras WSE-3
This system utilizes 32 × 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
Frequently asked questions
How much does the CS-3 Inference Cluster (32-node) cost?
Cerebras sells the CS-3 Inference Cluster (32-node) on quote through OEMs and resellers, and Flopper has no live rental listing for its WSE-3 accelerator to derive an hourly figure from. System rental pricing →
How much power does the CS-3 Inference Cluster (32-node) require?
Cerebras rates the CS-3 Inference Cluster (32-node) at 945 kW, which is the figure to plan power delivery and cooling around; per-GPU draw is listed on the accelerator's own page.
How many GPUs are in the CS-3 Inference Cluster (32-node)?
32. The CS-3 Inference Cluster (32-node) carries 32 Cerebras WSE-3 accelerators in a cluster form factor, and every throughput figure on this page is the total across them. WSE-3 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
Cerebras CS-3 Inference Cluster (32-node) specifications, generated from our database. Printable.
Vendor product page
Cerebras CS-3 Inference Cluster (32-node) documentation
Cerebras CS-4 datasheet
Cerebras • 2026-08-18
Typical Use Cases
The Cerebras CS-3 Inference Cluster (32-node) runs 32× WSE-3 GPUs over On-wafer fabric per engine, 214 Pb/s; cluster networking via dedicated NET cabinets.