RunPod GPU Cloud
AMD Instinct™ MI300X On-Demand
On-Demand AMD Instinct™ MI300X from $3.99/hr
RunPod works with Generative AI companies, Acedemic Institutions, and Enterprises
The world's fastest MI300X Cloud
Scale to hundreds of On-Demand Pods in minutes
Rapidly spin up AMD Instinct™ MI300X GPUs across multiple regions.
250ms Cold start times
As low as 100ms for certain workloads.
Pay by the minute for On-Demand Pods
Stop paying as soon as you spin down your workload.
Get StartedFrom $3.99/hr
Built for developer teams of all sizes
Startups
Enterprises
Secure Cloud
10Pb+ Multi-Region Network Storage
Access over 100Tb+ of network storage across NA and EU.
99.99% Uptime
Book a call to learn more about our reservation pricing.
Enterprise Grade Security
Best-in-class physical on-prem data center security.
AMD Instinct™ MI300X | H100 SXM | |
---|---|---|
On-Demand Pricing | $3.99/hr | $4.69/hr |
VRAM | 192 GB | 80 GB |
Memory Bandwidth | 5.3 TB/s | 3.4 TB/s |
FP64 | 81.7 TFLOPs | 67 TFLOPs |
FP32 | 163.4 TFLOPs | 67 TFLOPs |
FP16 | 2,610 TFLOPs | 1,979 TFLOPs |
FP8 | 5,220 TFLOPs | 3,958 TFLOPs |
Interconnect GPU to GPU | PCIe 128 GB/s | NVLink 900 GB/s PCIe 64 GB/s |
Network | 3.2 Tb/s | - |
The preferred platform for 100,000+ developers worldwide
Join our Discord community
Our dev community is happy to help with hands on support.
Check out our docs
Our docs provide a thorough dive into the RunPod platform.
Get started with tutorials
We have hundreds of tutorials for all types of workloads.
Total Requests
NaN
Flexible pricing on reservations
Discounts on 3, 6, 12, and 36-month reservations
Book a call with our team to learn more.
Zero fees for ingress / egress
No charges for data ingress and egress
Only pay for what you use, never more
RunPod only charges you for the resources used.
Secure Cloud | |
---|---|
On-Demand | $3.99/hr |
3-Month Reservation | Book a call |
6-Month Reservation | Book a call |
12-Month Reservation | Book a call |
36-Month Reservation | Book a call |