Runpod
AI developer cloud for building, deploying, and scaling AI workloads
Last reviewed Jul 16, 2026
Runpod is an AI developer cloud for building, deploying, and scaling AI workloads. It offers on‑demand GPU pods across 31 regions in two tiers — data‑center‑hosted Secure Cloud and lower‑cost Community Cloud — plus serverless endpoints and multi‑node clusters, with B300, B200, and H100/H200 alongside A100, L40S, RTX PRO 6000, and RTX classes.
Available GPUs
Hourly on-demand pricing. Click column headers to sort.
Prices last updated: October 26, 2026
| A100 PCIe 80GB | 1× | 1× | |||
| A100 PCIe 80GB | 1× | 1× | |||
| A100 SXM | 1× | 1× | |||
| A100 SXM | 1× | 1× | |||
| A100 SXM 40GB | 1× | 1× | |||
| A40 | 1× | 1× | |||
| B200 | 1× | 1× | |||
| B200 | 1× | 1× | |||
| H100 NVL | 1× | 1× | |||
| H100 NVL | 1× | 1× | |||
Runpod pricing by GPU
Configurations, price rank, and alternatives for one GPU at a time.
LLM API Pricing
Pay-per-token pricing. Prices shown per 1M tokens.
Prices last updated: October 20, 2026
| Model | Input/1M | Output/1M |
|---|---|---|
| $10.00 | $10.00 |
Runpod pricing by model
Input, output, and batch rates, plus alternatives, for one model at a time.
Pros & Cons
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
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
Flash
Run Python functions on remote GPUs directly from a local terminal
Public Endpoints
API access to pre-deployed image, video, audio, and text models with no infrastructure to manage
Hub
Deploy open-source AI models and templates on Runpod from a curated catalog
Compute Services
Pods
On‑demand single‑node GPU instances in Secure Cloud (data‑center hosted) or lower‑cost Community Cloud, with flexible templates and storage.
Serverless
Pay‑per‑second endpoints with automatic scaling for production AI/ML applications.
Instant Clusters
Fully managed multi‑node GPU clusters with high‑speed networking between nodes.
Pricing Options
| Option | Details |
|---|---|
| Secure Cloud | Pods hosted in vetted data centers; higher reliability tier |
| Community Cloud | Pods on vetted community hosts at lower rates than Secure Cloud |
| Spot | Interruptible capacity at reduced prices |
| Serverless | Pay-per-second billing for endpoint execution time, with automatic scaling and no charge when idle |
| Clusters | Per-hour or per-second billing for multi-node GPU clusters with no commitment, scaling up to 64 GPUs |
| Reserved Clusters | Discounted rates for 1, 3, 6, or 12+ month commitments on dedicated clusters with guaranteed availability and SLA-backed uptime; quoted by sales |
| Public Endpoints | Usage-based billing for pre-deployed model APIs, charged per request, per token, or per 1000 characters depending on the model |
Availability & Support
Regions
31 global regions worldwide
Support
Documentation site, support request form, public status page, and a community Discord
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
Frequently Asked Questions
What GPU types does Runpod offer?
Runpod offers various GPU types including RTX A5000, RTX A5000, RTX PRO 4000, RTX A6000, RTX A6000, A100 SXM, A100 SXM, RTX 5000 Ada, H100 SXM, H100 SXM, H200, H200, A100 PCIe 80GB, A100 PCIe 80GB, H100 NVL, H100 NVL, A100 SXM 40GB, L40, L40, RTX 5080, RTX 4080 SUPER, RTX 3080, RTX A2000, RTX 4000 Ada, RTX 4000 Ada, RTX 4080, RTX 4090, RTX 4090, RTX 3070, A40, RTX 6000 Ada, RTX 6000 Ada, HGX B300, HGX B300, RTX PRO 5000, RTX PRO 5000, RTX A4000, RTX A4000, H200 NVL, RTX 3080 Ti, RTX 3090 Ti, RTX A4500, RTX A4500, L40S, L40S, H100 PCIe, H100 PCIe, RTX 4070 Ti, RTX 5090, RTX 5090, B200, B200, Tesla V100, MI300X, L4, RTX PRO 6000, RTX PRO 6000, RTX 3090, RTX 3090, RTX PRO 4500. Check the pricing table above for current availability and pricing.
How do I get started with Runpod?
Create an account, Add payment method, Launch your first pod
What are Runpod's main advantages?
Runpod's main advantages include: Competitive pricing with pay-per-second billing, Wide variety of GPU options, Simple and intuitive interface.
What are Runpod's limitations?
Runpod's main limitations include: GPU availability can vary by region, Some features require technical knowledge.
Compare Providers
Find the best prices for the same GPUs and models from other providers
