GPU Cloud

Scalable infrastructure built for production.

Rent Cloud GPUs from $0.2/hour.

servers with a computer

GPU Instances
Secure / Community Cloud
Deploy container-based GPU instances that spin up in seconds using both public and private repositories.
Serverless GPUs
Secure Cloud
Pay per second serverless GPU computing brings autoscale to production, low cold-start, and security using our secure cloud.
AI Endpoints
Fully-Managed
Fully managed and scaled for any workload. Use our AI Endpoints for Dreambooth, Stable Diffusion, Whisper, and more.
Free Bandwidth
A100
80 GB


Starting at $1.79 / hr
$3.58
2x
$7.16
4x
$14.32
8x
RTX A6000
48 GB


Starting at $0.79 / hr
$1.58
2x
$3.16
4x
$6.32
8x
RTX A5000
24 GB


Starting at $0.44 / hr
$0.88
2x
$1.76
4x
$3.52
8x
RTX A4000
16 GB


Starting at $0.34 / hr
$0.68
2x
$1.36
4x
$2.72
8x
RTX 3090
24 GB


Starting at $0.44 / hr
$0.88
2x
$1.76
4x
$3.52
8x
and many more ...
Community
Cloud
Vetted Hosts
100Mbps - 1 Gbps Burst
Reliability > 98%
Many Compute Options
Rock-bottom Pricing
Secure
Cloud
Tier 3/4 Datacenters
Dedicated 1 Gbps Burst
Failover / Redundancy
Strict Privacy & Security
Enterprise Grade Hardware
Cloud Sync
Seamlessly download or upload pod data to any cloud storage.
0%
0 GiB / 2 GiB
500 MiB/s
runpodctl create pod --imageName=jupyter/tensorflow-notebook --gpuType="NVIDIA RTX A6000" --gpuCount=2 ...
pod "jqtsm6e3700cjz" created for $0.84 / hr
runpodctl get pod
ID
GPU
IMAGE NAME
STATUS
jqtsm6e3700cjz
2x RTX A6000
jupyter/tensorflow-notebook
RUNNING
runpodctl stop pod qmioq0dmowyfwz
pod "qmioq0dmowyfwz" stopped
runpodctl remove pod qmioq0dmowyfwz
pod "qmioq0dmowyfwz" removed