RunPod
RunPod offers on‑demand GPUs and instant multi‑node clusters across 30+ regions, with H100/H200 alongside A100, L40S, and RTX classes.
Key Features
- Secure Cloud GPUs
- Access to a wide range of GPU types with enterprise-grade security
- Pay-as-you-go
- Only pay for the compute time you actually use
- API Access
- Programmatically manage your GPU instances via REST API
- Fast cold-starts
- Pods typically ready in 20-30 s
- Hot-reload dev loop
- SSH & VS Code tunnels built-in
- Spot-to-on-demand fallback
- Automatic migration on pre-empt
Provider Comparison
Advantages
- Competitive pricing with pay-per-second billing
- Wide variety of GPU options
- Simple and intuitive interface
Limitations
- GPU availability can vary by region
- Some features require technical knowledge
Available GPUs
GPU Model↑ | Memory | Hourly Price |
|---|---|---|
A100 PCIE | 40GB | $1.89/hr |
A100 SXM | 80GB | $0.79/hr |
A40 | 48GB | $0.40/hr |
B200 | 192GB | $5.98/hr |
H100 | 80GB | $1.35/hr |
H200 | 141GB | $3.59/hr |
L40 | 40GB | $0.69/hr |
L40S | 48GB | $0.40/hr |
RTX 3090 | 24GB | $0.14/hr |
RTX 4090 | 24GB | $0.20/hr |
RTX 6000 Ada | 48GB | $0.40/hr |
RTX A4000 | 16GB | $0.09/hr |
RTX A5000 | 24GB | $0.11/hr |
RTX A6000 | 48GB | $0.25/hr |
Tesla V100 | 32GB | $0.17/hr |
Compute Services
Pods
On‑demand single‑node GPU instances with flexible templates and storage.
Instant Clusters
Spin up multi‑node GPU clusters in minutes with auto networking.
Getting Started
1
Create an account
Sign up for RunPod using your email or GitHub account
2
Add payment method
Add a credit card or cryptocurrency payment method
3
Launch your first pod
Select a template and GPU type to launch your first instance
Compare Providers
Find the best prices for the same GPUs from other providers
