How RunPod's per-second GPU cloud pricing works across Pods, Serverless, and Clusters, plus founding facts and who the platform is built for.
Category
AI Infrastructure & MLOps
Pricing
Usage-based, pay-per-second GPU billing with no subscription requirement, from From around $0.24 per GPU-hour depending on GPU type and cloud tier
Verified
Not yet
Last updated
July 19, 2026
Founded
2022
Headquarters
Moorestown, New Jersey, United States (remote-first team)
APIAI
Overview
RunPod is a GPU cloud platform focused specifically on AI and machine learning workloads. Founded in October 2022 by Zhen Lu and Pardeep Singh, the company is incorporated in Moorestown, New Jersey, but runs as a remote-first team spread across the US, Canada, Europe, and India. RunPod has grown quickly into a developer-focused alternative to hyperscale clouds, reporting service to hundreds of thousands of developers across nearly 200 countries with a lean employee base.
Key Features
RunPod splits its offering into Pods for dedicated GPU instances used in development and long-running jobs, Serverless for autoscaling inference workers billed only on active usage, and Clusters for multi-node distributed training. All three are billed per second with no minimum commitments and no egress fees, and Pods are further split into a lower-cost Community Cloud tier and a more isolated Secure Cloud tier depending on the sensitivity of the workload.
Pricing
RunPod pricing is entirely usage-based rather than subscription-based, with rates varying by GPU model and product line. Example on-demand Pod rates include roughly $2.99 per hour for an H100 SXM and $0.69 per hour for an RTX 4090, while Serverless inference workers range from about $0.69 to $4.55 per hour depending on GPU class. Storage is billed separately, from about $0.05 per GB per month for standard network storage up to $0.14 per GB per month for high-performance storage.
Key Features
Per-second GPU billing — Usage is billed down to the second rather than rounded to the hour, so short jobs only pay for time actually used.
Pods for dedicated instances — Dedicated GPU instances for development, training, and long-running jobs across Community and Secure Cloud tiers.
Serverless inference workers — Autoscaling GPU workers for inference workloads that scale to zero and bill only for active usage.
Multi-node Clusters — Reserved, multi-GPU cluster deployments for distributed training workloads at scale.
No egress fees — Data transfer out of RunPod is not charged separately, unlike many hyperscale cloud providers.
Wide GPU selection — Access to GPUs ranging from consumer-class RTX cards to data-center H100 and H200 accelerators.
Network and container storage — Separately priced persistent network storage and container disk options for model weights and data.
Global availability — Infrastructure reported to serve developers across 183 countries via a distributed data center footprint.
Pros & Cons
Pros
Per-second billing with no minimum commitments keeps costs predictable for short jobs
Wide range of GPU types from consumer cards to H100 and H200 accelerators
No egress fees, unlike many hyperscale cloud providers
Serverless tier autoscales inference workers to zero when idle
Lean, remote-first team has scaled to serve a large global developer base
Significantly cheaper than large hyperscalers for comparable GPU capacity
Cons
Community Cloud tier trades some security and reliability guarantees for lower price
Pricing spans multiple product lines, which can be confusing to compare at a glance
Specific GPU model availability can fluctuate with demand
Requires technical comfort with containers and APIs to get the most out of the platform
Support options are more limited without an enterprise-level relationship
Relatively young company founded in 2022, with less track record than legacy cloud providers
Pricing
Pods (Community Cloud) From around $0.69 per hour (e.g. RTX 4090) Per-second usage-based
Pods (Secure Cloud) From around $1.39 per hour (e.g. A100 PCIe) Per-second usage-based
Serverless From around $0.69 to $4.55 per hour depending on GPU Usage-based, billed for active worker time only
Clusters From around $1.79 per hour (e.g. A100 SXM) Per-hour usage-based, reserved options available
Frequently Asked Questions
What is RunPod used for
RunPod is a GPU cloud platform used for AI and machine learning workloads, including model training, fine-tuning, and inference.
How is RunPod priced
RunPod uses per-second usage-based billing across its Pods, Serverless, and Clusters products, with rates varying by GPU type and product line.
When was RunPod founded
RunPod was founded on October 31, 2022 by Zhen Lu and Pardeep Singh.
Where is RunPod headquartered
RunPod is incorporated in Moorestown, New Jersey, and operates as a remote-first team across the US, Canada, Europe, and India.
Does RunPod charge egress fees
No, RunPod advertises no egress fees for data transferred out of the platform.
What is the difference between Pods and Serverless on RunPod
Pods are dedicated GPU instances for development and long-running jobs, while Serverless autoscales inference workers based on traffic and bills only for active usage.
Does RunPod offer a free tier
RunPod does not advertise an ongoing free tier; billing is pay-as-you-go per second of GPU usage.
Who is RunPod built for
RunPod is built for developers, startups, and ML teams who need flexible, short-notice GPU access without long-term cloud contracts.