(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | gcp |
Server ID | 1740001 |
Name | ct5l-hightpu-1t |
Description | 24 vCPUs, 48 GB RAM, 1 Google TPUs |
Family | ct5l |
Hw Virt | - |
Status | active |
Observed At | 2026-10-06T15:11:33.973077 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Council Bluffs (US) / us-central1 | 0.3422 USD/h | 1.2 USD/h |
Eemshaven (NL) / europe-west4 | 0.5624 USD/h | 1.56 USD/h |
Processor
vCPUs | 24 |
CPU Allocation | Dedicated |
CPU Architecture | x86_64 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 48 GiB |
| GPU | |
|---|---|
GPU Count | 1 |
GPU Memory Min | 16 GiB |
GPU Memory Total | 16 GiB |
GPU Manufacturer | Google |
GPU Family | TPU |
GPU Model | v5e |
GPUs |
|
| STORAGE | |
|---|---|
Storage Size | 0 GB |
Storages |
| NETWORK | |
|---|---|
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
A dedicated compute instance featuring an integrated Google TPU v5e for cost-efficient machine learning training and inference workloads.
Google Cloud Platform ct5l-hightpu-1t is a specialized accelerator instance within the ct5l family, designed for machine learning and compute-heavy workloads. It is equipped with 24 dedicated x86_64 vCPUs and 48 GB of system memory, providing a 2.0 GB per core ratio. The core feature of this server is a single Google TPU v5e accelerator with 16 GB of VRAM, making it highly suitable for neural network training, fine-tuning, and inference. The instance does not include local storage, requiring external storage solutions for data persistence. This configuration offers a cost-efficient entry point for TPU-accelerated computing, balancing dedicated CPU resources with specialized hardware acceleration for machine learning pipelines.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | GPUs |
|---|---|---|---|
| ct5l-hightpu-4t | 112 | 192 GiB | 4 |
| ct5l-hightpu-8t | 224 | 384 GiB | 8 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | GPUs |
|---|