g4dn.4xlarge by Amazon Web Services

g4dn.4xlarge is a Graphics intensive [Instance store volumes] [Network and EBS optimized] Gen4 4xlarge server offered by Amazon Web Services with 16 vCPUs, 64 GiB of memory and 225 GB of storage. The pricing starts at 0.1768 USD per hour.
16 vCPU
64 GiB Memory
225 GB Storage
1 GPU
Spare SCore
13,550
(All-cores)
1,286
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g4dn.4xlarge
Description
Graphics intensive [Instance store volumes] [Network and EBS optimized] Gen4 4xlarge
Family
g4dn
Hw Virt
Status
active
Observed At
2026-10-24T22:03:23.044513

Availability

REGION / IDSPOTONDEMAND
Oregon (US) / us-west-2
0.483 USD/h 1.204 USD/h
Northern Virgina (US) / us-east-1
0.60238 USD/h 1.204 USD/h
Ohio (US) / us-east-2
0.6225 USD/h 1.204 USD/h
Stockholm (SE) / eu-north-1
0.209933 USD/h 1.277 USD/h
Mumbai (IN) / ap-south-1
0.499767 USD/h 1.325 USD/h
Quebec (CA) / ca-central-1
0.6167 USD/h 1.337 USD/h
Dublin (IE) / eu-west-1
0.610467 USD/h 1.342 USD/h
Paris (FR) / eu-west-3
0.446867 USD/h 1.408 USD/h
London (GB) / eu-west-2
0.402167 USD/h 1.409 USD/h
Milan (IT) / eu-south-1
0.24895 USD/h 1.41 USD/h
California (US) / us-west-1
0.71635 USD/h 1.445 USD/h
Seoul (KR) / ap-northeast-2
0.748167 USD/h 1.481 USD/h
Frankfurt (DE) / eu-central-1
0.5984 USD/h 1.505 USD/h
Sydney (AU) / ap-southeast-2
0.521133 USD/h 1.566 USD/h
Cape Town (ZA) / af-south-1
0.5734 USD/h 1.597 USD/h
Osaka (JP) / ap-northeast-3
0.4522 USD/h 1.625 USD/h
Tokyo (JP) / ap-northeast-1
0.745133 USD/h 1.625 USD/h
Singapore (SG) / ap-southeast-1
0.682533 USD/h 1.685 USD/h
Hong Kong (HK) / ap-east-1
0.319167 USD/h 1.854 USD/h
Sao Paulo (BR) / sa-east-1
0.209467 USD/h 2.046 USD/h

Processor

vCPUs
16
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
8
CPU Speed
2.5 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8259CL
CPU L1D Cache
32 KiB
CPU L1D Cache Total
256 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
256 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
8 MiB
CPU L3 Cache
36 MiB
CPU L3 Cache Total
36 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, xtopology, nonstop_tsc, cpuid, aperfmperf, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, pti, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, mpx, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, ida, arat, pku, ospke, avx512_vnni
Ecpus
10.5
Scalability
65.62

System Resources and Accelerators

MEMORY
Memory Amount
64 GiB
Memory Amount Actual
64 GiB
Memory Generation
DDR4
Memory Speed
3200 Mhz
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
16 GiB
Accelerator Memory Total
16 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Turing
Accelerator Model
T4
Accelerators
  • NVIDIA Turing Tesla T4 (Memory amount: 15360, Firmware version: 590.48.01, BIOS version: 90.04.96.00.02, Clock rate: 1590)
STORAGE
Storage Size
225 GB
Storage Type
nvme ssd
Storages
  • 225 GB nvme ssd
NETWORK
Network Speed Baseline
20 Gbps
Network Speed Max
25 Gbps
Network Storage Speed Baseline
4.75 Gbps
Network Storage Speed Max
4.75 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
0

CPU and System Topology

Server Description

A GPU-accelerated platform featuring an NVIDIA T4 GPU and NVMe storage, optimized for machine learning inference and graphics-intensive workloads.

GPU AcceleratedGeneral Purpose

Amazon Web Services g4dn.4xlarge is a GPU-accelerated instance designed for graphics-intensive and machine learning workloads. Running on the Nitro hypervisor, it features an Intel Xeon 8259CL x86_64 processor with 16 vCPUs, 8 physical cores, and 64.0 GB of DDR4 memory. The hardware includes a dedicated NVIDIA Turing T4 GPU with 16 GB of VRAM, a 225 GB NVMe SSD, and 20 Gbps baseline network bandwidth. Benchmark data shows a tradeoff: single-core CPU performance is in the bottom 25%, whereas GPU-driven LLM inference for small and medium models reaches the top 10% for prompt processing. This resource configuration makes the instance highly suitable for machine learning inference, graphics rendering, and workloads that leverage GPU acceleration rather than single-threaded CPU processing.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

0.76Score
Precomputed compound score for Cache Intensive workloads. A weighted average (geometric mean) of benchmark scores compared to their medians: score = ∏ (x_i / m_i)^(w_i / Σw). The score of 1.0 represents a synthetic baseline server with the median performance of each component benchmark; 0.5 means roughly half the performance; and 2.0 means twice the performance of that reference profile. Component weights: 50% Redis RPS (pipeline=1, SET), 20% Redis RPS (pipeline=16, SET), 10% PassMark Memory Mark (composite), 10% Memory bandwidth (read, 16 MB ~ L3), 10% PassMark single-thread CPU. Rationale for component selection: In-memory key-value store workload, mixing direct Redis performance metrics with memory speed and latency benchmarks, and single-core CPU performance profiles.
Component Score Weight Impact
Raw Reference Normalized Target Actual
Redis RPS (pipeline=1, SET) 1,355,587 ops/sec 1,919,289 ops/sec 0.74 50.00% 50.00% -14.00%
Redis RPS (pipeline=16, SET) 8,165,656 ops/sec 11,728,058 ops/sec 0.68 20.00% 20.00% -7.40%
PassMark Memory Mark (composite) 2,469 2,416 1 10.00% 10.00% 0.00%
Memory bandwidth (read, 16 MB ~ L3) 96,241 MB/sec 108,466 MB/sec 0.87 10.00% 10.00% -1.40%
PassMark single-thread CPU 1,816 Mops/s 2,365 Mops/s 0.75 10.00% 10.00% -2.80%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
5.7 GB/s
Node GPU-to-host bandwidth
6.6 GB/s
Node host-to-GPU bandwidth
6.3 GB/s
Single GPU slot
Per-GPU GPU-to-host bandwidth
6.6 GB/s
Per-GPU host-to-GPU bandwidth
6.3 GB/s
Per-GPU host-device latency
659 ns
Benchmark Environment
Driver Version
 535.183.01
CUDA Runtime Version
 12.6
nvbandwidth Version
 v0.5
GPU Count
 1
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,144

Geekbench Multi-Core

Score: 7,019

Passmark CPU Scores

BENCHMARKSCORE
Mark
 14,415
Compression
 199,877
Encryption
 6,242
Extended Instructions
 12,243
Floating Point Maths
 24,332
Integer Maths
 45,381
Physics
 1,580
Prime Numbers
 82
Single Threaded
 1,816
String Sorting
 23,520

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,469
Database Operations
 5,430
Memory Latency
 54
Memory Read Cached
 21,932
Memory Read Uncached
 10,349
Memory Write
 9,931

Stress-ng Raw Scores

Stress-ng Relative Multicore Performance

LLM Inference Speed for Prompt Processing

LLM Inference Speed for Text Generation

Static Web Server

Redis

Alternatives

Servers of the Same Family

INSTANCEvCPUsMEMORYAccelerators
g4dn.xlarge 416 GiB1
g4dn.2xlarge 832 GiB1
g4dn.8xlarge 32128 GiB1
g4dn.12xlarge 48192 GiB4
g4dn.16xlarge 64256 GiB1
g4dn.metal 96384 GiB8

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

g4dn.4xlarge FAQs