companies

RunPod

GPU cloud for AI builders, with on-demand pods, autoscaling serverless endpoints, and multi-node clusters billed by the second.

RunPod was founded in 2022 by Zhen Lu and Pardeep Singh after the two ran into the same wall every ML engineer hits: renting a GPU for a side project or a startup workload meant either overpaying a hyperscaler or wrestling with bare-metal providers built for enterprises. They started RunPod as a marketplace-style cloud that lets developers rent GPUs by the hour with none of the procurement overhead of AWS or GCP. The platform spans three main products: Pods, on-demand containers for training and interactive development across dozens of GPU types from RTX 4090s to H200s and B200s; Serverless, autoscaling GPU endpoints that spin down to zero between requests and use FlashBoot technology for fast cold starts; and Instant Clusters, multi-node setups with InfiniBand networking for distributed training. Billing runs per second across all three, with spot pricing for interruption-tolerant workloads sitting well below on-demand rates. RunPod is remote-first with team members across the US, Canada, Europe, and India, and closed a $100 million Series A led by Summit Partners that valued the company at roughly $1 billion, following an earlier seed round backed by Intel Capital and Dell Technologies Capital.

Founded

2022

Headquarters

San Francisco, United States

Sector

infrastructure