g5.48xlarge by Amazon Web Services

g5.48xlarge is a Graphics intensive Gen5 48xlarge server offered by Amazon Web Services with 192 vCPUs, 768 GiB of memory and 7.6 TB of storage. The pricing starts at 3.2479 USD per hour.
192 vCPU
768 GiB Memory
7.6 TB Storage
8 GPU
Spare SCore
73,696
(All-cores)
753
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g5.48xlarge
Description
Graphics intensive Gen5 48xlarge
Family
g5
Hw Virt
Status
active
Observed At
2026-10-27T03:24:27.062227

Availability

REGION / IDSPOTONDEMAND
Oregon (US) / us-west-2
6.322333 USD/h 16.288 USD/h
Northern Virgina (US) / us-east-1
10.54884 USD/h 16.288 USD/h
Ohio (US) / us-east-2
10.092133 USD/h 16.288 USD/h
Aragón (ES) / eu-south-2
16.7264 USD/h 17.1623 USD/h
Stockholm (SE) / eu-north-1
3.4099 USD/h 17.2756 USD/h
Quebec (CA) / ca-central-1
18.0852 USD/h 18.0852 USD/h
Dublin (IE) / eu-west-1
6.438533 USD/h 18.1823 USD/h
Tel Aviv (IL) / il-central-1
19.0915 USD/h 19.0914 USD/h
Mumbai (IN) / ap-south-1
16.1437 USD/h 19.5586 USD/h
United Arab Emirates / me-central-1
8.1791 USD/h 19.9795 USD/h
Seoul (KR) / ap-northeast-2
9.978667 USD/h 20.0281 USD/h
Frankfurt (DE) / eu-central-1
20.3681 USD/h 20.3681 USD/h
London (GB) / eu-west-2
20.6757 USD/h 20.6757 USD/h
Paris (FR) / eu-west-3
20.6757 USD/h 20.6757 USD/h
Sydney (AU) / ap-southeast-2
21.1776 USD/h 21.1776 USD/h
Jakarta (ID) / ap-southeast-3
11.0711 USD/h 22.7967 USD/h
Tokyo (JP) / ap-northeast-1
21.2602 USD/h 23.6225 USD/h
Hong Kong (HK) / ap-east-1
21.0989 USD/h 25.0764 USD/h
Sao Paulo (BR) / sa-east-1
22.59695 USD/h 27.6864 USD/h

Processor

vCPUs
192
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
96
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
3 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
384 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
97.8
Scalability
50.94

System Resources and Accelerators

MEMORY
Memory Amount
768 GiB
Memory Amount Actual
768 GiB
Memory Generation
DDR4
Memory Speed
3200 Mhz
ACCELERATORS
Accelerator Count
8
Accelerator Memory Min
22 GiB
Accelerator Memory Total
179 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: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
  • NVIDIA Ampere NVIDIA A10G (Memory amount: 23028, Firmware version: 590.48.01, BIOS version: 94.02.91.00.07, Clock rate: 1710)
STORAGE
Storage Size
7600 GB
Storage Type
nvme ssd
Storages
  • 3800 GB nvme ssd
  • 3800 GB nvme ssd
NETWORK
Network Speed Baseline
100 Gbps
Network Speed Max
100 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 featuring eight accelerators, massive local NVMe storage, and high-bandwidth networking for parallelized workloads.

GPU AcceleratedCompute OptimizedStorage & Database

Amazon Web Services g5.48xlarge is a high-density, GPU-accelerated server instance built on the Nitro hypervisor. It features 192 vCPUs from an AMD EPYC 7R32 processor running at 3.3 GHz, paired with 768.0 GB of DDR4 memory and 7600 GB of local NVMe SSD storage. Acceleration is driven by eight NVIDIA Ampere A10G GPUs with 179 GB of total VRAM. Benchmark data indicates top-tier performance in multi-threaded operations, Redis database transactions, and large language model inference across various model sizes, contrasting with lower single-core performance. With a baseline network bandwidth of 100 Gbps, this instance is highly suited for parallelized workloads, including LLM prompt processing, ray tracing, and intensive database operations.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

4.5Score
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) 13,744,651 ops/sec 1,918,343 ops/sec 7.4 50.00% 50.00% +170.00%
Redis RPS (pipeline=16, SET) 93,031,803 ops/sec 11,704,734 ops/sec 7.8 20.00% 20.00% +51.00%
PassMark Memory Mark (composite) 2,885 2,416 1.2 10.00% 10.00% +1.80%
Memory bandwidth (read, 16 MB ~ L3) 243,165 MB/sec 108,420 MB/sec 2.2 10.00% 10.00% +8.20%
PassMark single-thread CPU 2,099 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
60.8 GB/s
Node GPU-to-host bandwidth
80.9 GB/s
Node host-to-GPU bandwidth
96.5 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
982 ns
Benchmark Environment
Driver Version
 535.183.01
CUDA Runtime Version
 12.6
nvbandwidth Version
 v0.5
GPU Count
 8
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,307

Geekbench Multi-Core

Score: 14,102

Passmark CPU Scores

BENCHMARKSCORE
Mark
 94,547
Compression
 2,724,708
Encryption
 202,750
Extended Instructions
 155,611
Floating Point Maths
 385,753
Integer Maths
 646,437
Physics
 13,100
Prime Numbers
 863
Single Threaded
 2,099
String Sorting
 349,079

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,885
Database Operations
 22,765
Memory Latency
 52
Memory Read Cached
 24,184
Memory Read Uncached
 11,945
Memory Write
 12,941

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
g5.24xlarge 96384 GiB4

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

g5.48xlarge FAQs