g5.xlarge by Amazon Web Services

g5.xlarge is a Graphics intensive Gen5 xlarge server offered by Amazon Web Services with 4 vCPUs, 16 GiB of memory and 250 GB of storage. The pricing starts at 0.1408 USD per hour.
4 vCPU
16 GiB Memory
250 GB Storage
1 GPU
Spare SCore
1,533
(All-cores)
767
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g5.xlarge
API Reference Object
-
Description
Graphics intensive Gen5 xlarge
Family
g5
Hw Virt
Status
active
Observed At
2026-10-17T07:16:38.113174

Availability

REGION / IDSPOTONDEMAND
Northern Virgina (US) / us-east-1
0.4634 USD/h 1.006 USD/h
Oregon (US) / us-west-2
0.5066 USD/h 1.006 USD/h
Ohio (US) / us-east-2
0.540267 USD/h 1.006 USD/h
Aragón (ES) / eu-south-2
0.402833 USD/h 1.06 USD/h
Stockholm (SE) / eu-north-1
0.33005 USD/h 1.067 USD/h
Quebec (CA) / ca-central-1
0.49165 USD/h 1.117 USD/h
Dublin (IE) / eu-west-1
0.6706 USD/h 1.123 USD/h
Tel Aviv (IL) / il-central-1
0.792067 USD/h 1.1791 USD/h
Mumbai (IN) / ap-south-1
0.6332 USD/h 1.208 USD/h
United Arab Emirates / me-central-1
0.5052 USD/h 1.234 USD/h
Seoul (KR) / ap-northeast-2
0.579833 USD/h 1.237 USD/h
Frankfurt (DE) / eu-central-1
0.642367 USD/h 1.258 USD/h
Paris (FR) / eu-west-3
0.5006 USD/h 1.277 USD/h
London (GB) / eu-west-2
0.6439 USD/h 1.277 USD/h
Sydney (AU) / ap-southeast-2
0.652 USD/h 1.308 USD/h
Jakarta (ID) / ap-southeast-3
0.44705 USD/h 1.408 USD/h
Tokyo (JP) / ap-northeast-1
0.73385 USD/h 1.459 USD/h
Hong Kong (HK) / ap-east-1
0.42325 USD/h 1.5488 USD/h
Sao Paulo (BR) / sa-east-1
0.3631 USD/h 1.71 USD/h

Processor

vCPUs
4
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
2
CPU Speed
3.3 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7R32
CPU L1D Cache
32 KiB
CPU L1D Cache Total
64 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
64 KiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
1 MiB
CPU L3 Cache
8 MiB
CPU L3 Cache Total
8 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, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, nonstop_tsc, cpuid, extd_apicid, aperfmperf, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, topoext, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, clzero, xsaveerptr, rdpru, wbnoinvd, arat, npt, nrip_save, rdpid
Ecpus
2
Scalability
50

System Resources and Accelerators

MEMORY
Memory Amount
16 GiB
Memory Amount Actual
16 GiB
Memory Generation
DDR4
Memory Speed
3200 Mhz
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
22 GiB
Accelerator Memory Total
22 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Ampere
Accelerator Model
A10G
Accelerators
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.75.00.01, Clock rate: 1710)
STORAGE
Storage Size
250 GB
Storage Type
nvme ssd
Storages
  • 250 GB nvme ssd
NETWORK
Network Speed Baseline
2.5 Gbps
Network Speed Max
10 Gbps
Network Storage Speed Baseline
0.7 Gbps
Network Storage Speed Max
3.5 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
0

CPU and System Topology

Server Description

An accelerator-focused instance pairing dedicated AMD EPYC processing with an NVIDIA Ampere GPU for high-efficiency machine learning inference.

GPU AcceleratedGeneral Purpose

Amazon Web Services g5.xlarge is an x86_64 server featuring 4 dedicated vCPUs from an AMD EPYC 7R32 processor, 16.0 GB of DDR4 RAM, and a 250 GB NVMe SSD. It is equipped with a single NVIDIA Ampere A10G GPU with 22 GB of VRAM, running on the Nitro hypervisor with a 2 Gbps baseline network speed. Performance benchmarks highlight a significant trade-off: while CPU and memory bandwidth tests show weak to poor results, the GPU delivers top-tier performance for LLM inference on small and medium models. This makes the instance highly cost-efficient for GPU-centric tasks but less suitable for general CPU-bound processing. It is best fit for low-latency LLM inference, machine learning evaluation, and graphics-intensive workloads.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

0.26Score
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) 337,064 ops/sec 1,917,050 ops/sec 0.18 50.00% 50.00% -58.00%
Redis RPS (pipeline=16, SET) 2,312,964 ops/sec 11,661,818 ops/sec 0.19 20.00% 20.00% -28.00%
PassMark Memory Mark (composite) 1,990 2,417 0.83 10.00% 10.00% -1.80%
Memory bandwidth (read, 16 MB ~ L3) 35,916 MB/sec 108,361 MB/sec 0.33 10.00% 10.00% -10.00%
PassMark single-thread CPU 2,137 Mops/s 2,365 Mops/s 0.88 10.00% 10.00% -1.30%

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,333

Geekbench Multi-Core

Score: 3,052

Passmark CPU Scores

BENCHMARKSCORE
Mark
 5637
Compression
 60807
Encryption
 4271
Extended Instructions
 3620
Floating Point Maths
 8057
Integer Maths
 13468
Physics
 608
Prime Numbers
 35
Single Threaded
 2137
String Sorting
 8712

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 1990
Database Operations
 1431
Memory Latency
 50
Memory Read Cached
 24464
Memory Read Uncached
 12383
Memory Write
 12699

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
g5.2xlarge 832 GiB1
g5.4xlarge 1664 GiB1
g5.8xlarge 32128 GiB1
g5.12xlarge 48192 GiB4
g5.16xlarge 64256 GiB1
g5.24xlarge 96384 GiB4
g5.48xlarge 192768 GiB8

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

g5.xlarge FAQs