Providers

We monitor real-time GPU rental prices from these cloud providers and present them through our index.

Vast.ai

Vast.ai is a decentralized marketplace that connects GPU owners with developers seeking compute. Users can pick and choose from various hardware configurations—ranging from consumer RTX cards to high-end V100 or A100 GPUs—often at lower cost points.

Akash Network

Akash provides a decentralized marketplace for compute, using blockchain technology to match GPU resource providers with developers. This peer-to-peer approach can result in lower costs and a wider variety of available GPU models, though availability may fluctuate.

Cudo Compute

Cudo Compute, addresses diverse GPU workloads, spanning AI model training, HPC tasks, and even crypto mining. It leverages container technologies and offers monitoring tools initially developed for mining but equally effective for tracking AI job metrics.

Hyperstack

Hyperstack is a newer service that focuses on enterprise-grade GPU performance for AI, HPC, and real-time inference workloads. It aims to streamline complex deployments without sacrificing raw power.

Massed Compute

Massed Compute is geared toward GPU-intensive workloads such as AI/ML training, data analytics, and scientific simulations. It differentiates itself with a clean provisioning interface and support for a range of NVIDIA GPUs, from T4 instances for lighter tasks to A100 instances for heavy-duty machine learning.

Paperspace

Paperspace offers both raw GPU instances and a specialized machine learning development suite called Gradient. Gradient supplies hosted Jupyter notebooks, automated pipelines, and containerized environments for training, making it accessible to a wide range of developers.

Runpod

RunPod focuses on rapid deployment of GPU instances, appealing to both researchers prototyping small models and developers training large AI workloads. It uses container-based technology, so you can easily port your Docker images or select from community-provided ML frameworks.

Lambda Labs

Lambda specializes in GPU infrastructure optimized for machine learning, making it a strong choice for AI developers and research labs. Their services come with Lambda Stack—a pre-configured environment that includes commonly used deep learning frameworks, CUDA, and NVIDIA drivers.