(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | gcp |
Server ID | 999920000 |
Name | g4-standard-48 |
API Reference Object | - |
Description | Graphics Optimized: 1 NVIDIA RTX PRO 6000 GPU, 48 vCPUs, 180GB RAM |
Family | g4 |
Hw Virt | - |
Status | active |
Observed At | 2026-10-17T07:38:24.304406 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Council Bluffs (US) / us-central1 | 1.6294 USD/h | 4.4999 USD/h |
The Dalles (US) / us-west1 | 1.6675 USD/h | 4.4999 USD/h |
Moncks Corner (US) / us-east1 | 1.6675 USD/h | 4.4999 USD/h |
Ashburn (US) / us-east4 | 1.7937 USD/h | 4.4999 USD/h |
Dallas (US) / us-south1 | 1.8022 USD/h | 4.4999 USD/h |
Hamina (FI) / europe-north1 | 2.0543 USD/h | 4.9499 USD/h |
Salt Lake City (US) / us-west3 | 2.0642 USD/h | 4.9499 USD/h |
Eemshaven (NL) / europe-west4 | 2.2217 USD/h | 4.9499 USD/h |
St. Ghislain (BE) / europe-west1 | 2.2292 USD/h | 4.9499 USD/h |
Las Vegas (US) / us-west4 | 2.3154 USD/h | 4.9499 USD/h |
Columbus (US) / us-east5 | 2.5243 USD/h | 4.9499 USD/h |
Berlin (DE) / europe-west10 | 2.3515 USD/h | 5.3991 USD/h |
Dammam (SA) / me-central2 | 2.4658 USD/h | 5.3991 USD/h |
London (GB) / europe-west2 | 2.2385 USD/h | 5.3999 USD/h |
Jurong West (SG) / asia-southeast1 | 2.2951 USD/h | 5.3999 USD/h |
Milan (IT) / europe-west8 | 2.4935 USD/h | 5.3999 USD/h |
Mumbai (IN) / asia-south1 | 2.5079 USD/h | 5.8496 USD/h |
Jakarta (ID) / asia-southeast2 | 2.513 USD/h | 5.8496 USD/h |
Tokyo (JP) / asia-northeast1 | 2.2247 USD/h | 5.8499 USD/h |
Melbourne (AU) / australia-southeast2 | 2.2247 USD/h | 5.8499 USD/h |
Delhi (IN) / asia-south2 | 2.5556 USD/h | 5.8499 USD/h |
Changhua County (TW) / asia-east1 | 2.7358 USD/h | 5.8499 USD/h |
Processor
vCPUs | 48 |
CPU Allocation | Dedicated |
CPU Architecture | x86_64 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 180 GiB |
| ACCELERATORS | |
|---|---|
Accelerator Count | 1 |
Accelerator Memory Min | 96 GiB |
Accelerator Memory Total | 96 GiB |
Accelerator Type | GPU |
Accelerator Manufacturer | NVIDIA |
Accelerator Family | Blackwell |
Accelerator Model | RTX PRO 6000 |
Accelerators |
|
| STORAGE | |
|---|---|
Storage Size | 0 GB |
Storages |
| NETWORK | |
|---|---|
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
An x86_64 server featuring a ninety-six gigabyte Ada Lovelace GPU and forty-eight dedicated virtual cores for graphics-intensive workloads.
Google Cloud Platform g4-standard-48 is an x86_64 architecture server optimized for graphics-intensive workloads. The instance is configured with 48 dedicated vCPUs and 180.0 GB of system memory, which equates to 3.75 GB of RAM per core. Acceleration is provided by a single NVIDIA RTX Pro 6000 GPU from the Ada Lovelace architecture family, featuring 96 GB of dedicated VRAM. This server does not include local storage, meaning all storage requirements must be fulfilled through network-attached volumes. It also includes zero complimentary public IPv4 addresses. The high VRAM capacity relative to the system memory and CPU cores makes this server suitable for workloads such as 3D rendering, visualization, and machine learning inference where large model weights or textures must reside directly in the GPU frame buffer.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | Accelerators |
|---|---|---|---|
| g4-standard-6 | 6 | 22 GiB | ⅛ |
| g4-standard-12 | 12 | 45 GiB | ¼ |
| g4-standard-24 | 24 | 90 GiB | ½ |
| g4-standard-96 | 96 | 360 GiB | 2 |
| g4-standard-192 | 192 | 720 GiB | 4 |
| g4-standard-384 | 384 | 1440 GiB | 8 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | Accelerators |
|---|