AWS EC2 Instances›GPU Instance›g5›g5.48xlarge

Compute

vCPUs192
Memory768 GiB
Mem / vCPU4.00 GiB
ProcessorEPYC 7R32
Clock speed2.8 GHz
Architecturex86_64
GPU8
GPU Wattage41 W
GPU ArchitectureAmpere, Ampere, Ampere, Ampere, Ampere, Ampere, Ampere, Ampere
Video Memory192 GiB
FPGA0
ffmpeg FPS110
CoreMark / sec3,614,458

Storage

Type2 × 3,800 GB NVMe SSD
Bandwidth19,000 Mbps
Throughput2375 MB/s
IOPS80,000
Base Bandwidth19,000 Mbps
Base Throughput2375 MB/s
Base IOPS80,000
Devices2
NVMe DriveYes
SSDYes
Swap PartitionNo
Local Read IOPS625,000
Local Write IOPS312,500

Networking

Performance100 Gigabit
Max ENIs7
IPv6Yes
GPUNVIDIA A10G
Placement GroupSupported
TrunkingYes
Branch Interface115
IPs / ENI50
Baseline Bandwidth100 Gbps

NUMA Architecture

Uses NUMAYes
Node Count2
Max Distance32
Cores / Node48
Threads / Node96
Memory / Node382,923.5 MB
L3 Cache / Node192 MB
L3 Cache SharedYes

AWS Metadata

Instance typeg5.48xlarge
Full nameG5 Graphics and Machine Learning GPU 48xlarge
FamilyGPU instance
Variantg5
GenerationCurrent
Bare MetalNo
EMR supportYes
DedicatedSupported
Introduced2021-11-12
Family sizes

g5.48xlarge

AWS's g5.48xlarge is a GPU instance with 192 vCPU and 768 GiB. Rackspace Spot has no equivalent to offer: 8 × NVIDIA A10G wired together behind a high-speed interconnect is one tightly coupled machine, not 8 independent nodes, and that interconnect is usually what the instance was bought for.

Rackspace Spot's largest VM tops out at 16 vCPU and 128 GiB, and even that is not the missing piece here. The table below prices what AWS charges; the honest answer on the Rackspace Spot side is that this workload stays where it is.

Server comparison caveats
No equivalent

The AWS g5.48xlarge is 8 Ampere, Ampere, Ampere, Ampere, Ampere, Ampere, Ampere, Ampere GPUs (192 GiB VRAM) wired together inside one chassis. The interconnect is what you bought (that is what makes it one accelerator with pooled memory rather than several machines), and Rackspace Spot has nothing that reproduces it. The Rackspace Spot GPU class is built on the NVIDIA A30 (Ampere, 24 GB VRAM), and separate nodes of it on ordinary datacentre networking are not a substitute for a single interconnect-joined node.

Split the job across independent single-GPU nodes only if it is data-parallel with no per-step synchronisation and the model fits one card. Model-parallel work, or anything synchronising gradients every step, does not survive the move. Rackspace Spot sells x86_64 servers up to 16 vCPU, plus one NVIDIA A30 GPU class.

Pricing by region

1 hr

Regions are Rackspace Spot locations.

RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace on-demandRackspace SpotOn-demand price savingsSpot price savings
US East, N. Virginia$16.288$10.565$10.261$7.036not offerednot offeredn/an/a
US Central, Chicagono AWS regionn/an/an/anot offerednot offeredn/an/a
US Central, Dallasno AWS regionn/an/an/anot offerednot offeredn/an/a
Hong Kong$25.076$12.538$15.798$10.833not offerednot offeredn/an/a
Sydney, Australia$21.178$21.178$13.342$9.149not offerednot offeredn/an/a
London, UK$20.676$19.406$13.026$8.932not offerednot offeredn/an/a
US West, San Josenot sold heren/an/an/anot offerednot offeredn/an/a
US Central, Dallas 2no AWS regionn/an/an/anot offerednot offeredn/an/a
US East, N. Virginia 2$16.288$10.565$10.261$7.036not offerednot offeredn/an/a

Prices updated 17 Nov 2026.

When to use it

Good for
  • ●Model inference and embedding generation
  • ●Video transcoding and image processing
  • ●Machine learning experimentation and fine-tuning
  • ●Workloads that fit inside 768 GiB of RAM
  • ●Jobs that want 2 × 3,800 GB of local NVMe scratch space
  • ●Network-heavy services (AWS rates this size at 100 Gigabit)
Avoid if
  • ●Starting a new deployment. g6.48xlarge is the newer generation at this size.
  • ●The working set needs more than 768 GiB of RAM
  • ●The work is single-threaded. Most of these 192 vCPU would sit idle.
  • ●Anything the CPU can do fast enough. GPU hours are the most expensive on the price list.
  • ●More than 4.00 GiB of RAM per vCPU is needed, since that ratio is fixed for the whole family
Rackspace Spot sells spare capacity at market prices. You pay what the market decides, not a list price.
  • Rackspace Spot sells spare capacity at market prices. It is the cheapest compute on the internet.
  • Prices are set by supply and demand. You pay what the market decides.
  • Compute is delivered as fully managed Kubernetes. All infrastructure is taken care of.
  • Egress costs $0.01/GB after the free allowance. AWS charges $0.09/GB.
Build your Cloudspace

Frequently asked questions

How much does g5.48xlarge cost?

The AWS g5.48xlarge costs $16.29/hr on-demand in US East (N. Virginia), about $11,890/mo running 24/7, or $10.57/hr at the AWS Spot price. Rackspace Spot also sells in Chicago and Dallas, where AWS operates no region at all, so those rows have no AWS price to compare against.

How much is g5.48xlarge per month?

At 730 hours a month, g5.48xlarge in US East (N. Virginia) is $11,890/mo on AWS on-demand, or $7,713/mo at the AWS Spot price.

Does Rackspace Spot have an equivalent to g5.48xlarge?

No, and it is worth saying so plainly: 8 × NVIDIA A10G wired together behind a high-speed interconnect is one tightly coupled machine, not 8 independent nodes, and that interconnect is usually what the instance was bought for. Rackspace Spot sells market-priced CPU and GPU capacity, which is a good fit for the batch, web and inference work around this instance, but not for g5.48xlarge itself.

Is g5.48xlarge still worth using?

For workloads already running on it, yes: it works and it is priced accordingly. For anything new, g6.48xlarge is the current answer and measures roughly 6% more CoreMark throughput at the same vCPU count. 2 newer generations of the g family have shipped since (g6 and g7), which is the signal that matters for infrastructure you expect to keep running.

What is the difference between g5 and g6?

g6 is the newer generation. It runs on AMD EPYC 7R13 Processor against g5's AMD EPYC 7R32. On the CoreMark figures in this dataset, g6.48xlarge scores 3,842,690 per second against g5.48xlarge's 3,614,458. On-demand in US East (N. Virginia) the newer size is 18% cheaper. Neither generation changes the Rackspace Spot side: both map to the same Rackspace Spot servers.

What is a cheaper alternative to g5.48xlarge?

Not on Rackspace Spot, for the reason above. Within AWS the usual levers are a smaller size in the same family, a reserved or savings-plan commitment, or AWS Spot capacity if the work can tolerate interruption.