(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | gcp |
Server ID | 1741017 |
Name | ct6e-standard-8t |
Description | 360 vCPUs, 1440 GB RAM, 8 Google V6 lite TPUs |
Family | ct6e |
Hw Virt | - |
Status | active |
Observed At | 2026-10-24T02:21:42.274065 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Ashburn (US) / us-east4 | 1.944 USD/h | 21.6 USD/h |
Council Bluffs (US) / us-central1 | 4.4371 USD/h | 21.6 USD/h |
The Dalles (US) / us-west1 | 4.4371 USD/h | 21.6 USD/h |
Moncks Corner (US) / us-east1 | 4.4371 USD/h | 21.6 USD/h |
Dallas (US) / us-south1 | 5.0035 USD/h | 21.6 USD/h |
Santiago (CL) / southamerica-west1 | 8.64 USD/h | 21.6 USD/h |
Columbus (US) / us-east5 | 12.96 USD/h | 21.6 USD/h |
Jurong West (SG) / asia-southeast1 | 2.1384 USD/h | 23.76 USD/h |
Eemshaven (NL) / europe-west4 | 14.256 USD/h | 23.76 USD/h |
Mumbai (IN) / asia-south1 | 11.9289 USD/h | 25.92 USD/h |
Tokyo (JP) / asia-northeast1 | 12.1175 USD/h | 25.92 USD/h |
Processor
vCPUs | 360 |
CPU Allocation | Dedicated |
CPU Architecture | x86_64 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 1 TB |
| ACCELERATORS | |
|---|---|
Accelerator Count | 8 |
Accelerator Memory Min | 32 GiB |
Accelerator Memory Total | 256 GiB |
Accelerator Type | TPU |
Accelerator Manufacturer | Google |
Accelerator Family | TPU |
Accelerator Model | v6e |
Accelerators |
|
| STORAGE | |
|---|---|
Storage Size | 0 GB |
Storages |
| NETWORK | |
|---|---|
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
A high-density accelerator instance featuring three hundred sixty dedicated vCPUs and eight integrated TPUs for demanding machine learning workloads.
Google Cloud Platform ct6e-standard-8t is a specialized accelerator instance built on the x86_64 architecture, featuring 360 dedicated vCPUs and 1440.0 GB of system memory. Designed for high-throughput machine learning workloads, the server integrates eight Google TPU v6e accelerators with a total of 256 GB of VRAM. It provides a balanced memory-to-core ratio of 4.0 GB per vCPU but lacks local storage, requiring external network storage solutions for data persistence. This high-density hardware profile makes the instance highly cost-efficient for consolidating large-scale deep learning training and inference pipelines. The ct6e-standard-8t is best suited for data science jobs, machine learning model training, and large-scale AI inference workloads.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | Accelerators |
|---|---|---|---|
| ct6e-standard-1t | 44 | 176 GiB | 1 |
| ct6e-standard-4t | 180 | 720 GiB | 4 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | Accelerators |
|---|