g5.24xlarge by Amazon Web Services

g5.24xlarge is a Graphics intensive Gen5 24xlarge server offered by Amazon Web Services with 96 vCPUs, 384 GiB of memory and 3.8 TB of storage. The pricing starts at 2.2609 USD per hour.
96 vCPU
384 GiB Memory
3.8 TB Storage
4 GPU
Spare SCore
36,855
(All-cores)
753
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g5.24xlarge
Description
Graphics intensive Gen5 24xlarge
Family
g5
Hw Virt
Status
active
Observed At
2026-10-26T14:19:35.637420

Availability

REGION / IDSPOTONDEMAND
Oregon (US) / us-west-2
3.259033 USD/h 8.144 USD/h
Northern Virgina (US) / us-east-1
4.08126 USD/h 8.144 USD/h
Ohio (US) / us-east-2
4.016667 USD/h 8.144 USD/h
Aragón (ES) / eu-south-2
8.4006 USD/h 8.5811 USD/h
Stockholm (SE) / eu-north-1
2.4159 USD/h 8.6378 USD/h
Quebec (CA) / ca-central-1
6.4216 USD/h 9.0426 USD/h
Dublin (IE) / eu-west-1
3.414433 USD/h 9.0912 USD/h
Tel Aviv (IL) / il-central-1
9.5457 USD/h 9.5457 USD/h
Mumbai (IN) / ap-south-1
8.8921 USD/h 9.7793 USD/h
United Arab Emirates / me-central-1
4.0896 USD/h 9.9898 USD/h
Seoul (KR) / ap-northeast-2
4.9348 USD/h 10.014 USD/h
Frankfurt (DE) / eu-central-1
10.08065 USD/h 10.184 USD/h
London (GB) / eu-west-2
10.3379 USD/h 10.3379 USD/h
Paris (FR) / eu-west-3
10.3379 USD/h 10.3379 USD/h
Sydney (AU) / ap-southeast-2
10.5888 USD/h 10.5888 USD/h
Jakarta (ID) / ap-southeast-3
5.5748 USD/h 11.3984 USD/h
Tokyo (JP) / ap-northeast-1
10.6301 USD/h 11.8112 USD/h
Hong Kong (HK) / ap-east-1
12.5382 USD/h 12.5382 USD/h
Sao Paulo (BR) / sa-east-1
7.68505 USD/h 13.8432 USD/h

Processor

vCPUs
96
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
48
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
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
24 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
192 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, monitor, 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, perfctr_core, 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
48.9
Scalability
50.94

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
384 GiB
Memory Generation
DDR4
Memory Speed
3200 Mhz
ACCELERATORS
Accelerator Count
4
Accelerator Memory Min
22 GiB
Accelerator Memory Total
89 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)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.75.00.01, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.75.00.01, Clock rate: 1710)
  • 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
3800 GB
Storage Type
nvme ssd
Storages
  • 3800 GB nvme ssd
NETWORK
Network Speed Baseline
50 Gbps
Network Speed Max
50 Gbps
Network Storage Speed Baseline
19 Gbps
Network Storage Speed Max
19 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
0

CPU and System Topology

Server Description

A high-density GPU instance combining four NVIDIA accelerators with local NVMe storage for advanced machine learning inference and parallelized workloads.

GPU AcceleratedCompute OptimizedStorage & Database

Amazon Web Services g5.24xlarge is a GPU-accelerated instance featuring 96 dedicated vCPUs powered by an AMD EPYC 7R32 processor at 3.3 GHz, 384.0 GB of DDR4 RAM, and 3800 GB of local NVMe SSD storage. It integrates four NVIDIA Ampere A10G GPUs with 89 GB of total VRAM, supported by a 50 Gbps baseline network. While single-core CPU benchmarks rank in the bottom 10%, the instance delivers top-tier performance in multi-threaded and GPU-driven workloads. It achieves exceptional results in LLM inference prompt processing and text generation across various model sizes, alongside strong performance in cryptography, ray tracing, and software compilation. The g5.24xlarge is highly efficient for parallelized workloads, making it best suited for machine learning inference, image processing, and graphics-intensive applications.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

2.6Score
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) 7,456,335 ops/sec 1,919,289 ops/sec 3.9 50.00% 50.00% +97.00%
Redis RPS (pipeline=16, SET) 47,622,894 ops/sec 11,728,058 ops/sec 4 20.00% 20.00% +32.00%
PassMark Memory Mark (composite) 2,901 2,416 1.2 10.00% 10.00% +1.80%
Memory bandwidth (read, 16 MB ~ L3) 121,321 MB/sec 108,466 MB/sec 1.1 10.00% 10.00% +0.96%
PassMark single-thread CPU 2,093 Mops/s 2,365 Mops/s 0.88 10.00% 10.00% -1.30%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
44.0 GB/s
Node GPU-to-host bandwidth
52.9 GB/s
Node host-to-GPU bandwidth
53.9 GB/s
Single GPU slot
Per-GPU GPU-to-host bandwidth
13.2 GB/s
Per-GPU host-to-GPU bandwidth
13.5 GB/s
Per-GPU host-device latency
924 ns
Benchmark Environment
Driver Version
 535.183.01
CUDA Runtime Version
 12.6
nvbandwidth Version
 v0.5
GPU Count
 4
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,310

Geekbench Multi-Core

Score: 14,734

Passmark CPU Scores

BENCHMARKSCORE
Mark
 68,547
Compression
 1,368,947
Encryption
 101,562
Extended Instructions
 78,157
Floating Point Maths
 193,153
Integer Maths
 323,582
Physics
 6,710
Prime Numbers
 434
Single Threaded
 2,093
String Sorting
 177,391

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,901
Database Operations
 21,621
Memory Latency
 52
Memory Read Cached
 24,182
Memory Read Uncached
 12,674
Memory Write
 13,425

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.xlarge 416 GiB1
g5.2xlarge 832 GiB1
g5.4xlarge 1664 GiB1
g5.8xlarge 32128 GiB1
g5.12xlarge 48192 GiB4
g5.16xlarge 64256 GiB1
g5.48xlarge 192768 GiB8

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

g5.24xlarge FAQs