GPU marketplacesOnline
AutoDL
AutoDL is a Chinese GPU rental cloud offering exclusive-GPU container instances (RTX 4090, A100, A800, H800, H20, L20 and more) with pay-as-you-go billing to…
Compare GPU marketplaces
Read the source for billing, capacity and GPU details.
Results
23 GPU cloud providers
GPU neocloudsOnline
Mithril (formerly Foundry / ML Foundry)
Self-serve GPU cloud from Foundry Technologies offering NVIDIA H100 and H200 instances via market-priced spot bids and flexible reservations, with API, CLI and…
GPU marketplacesOnline
Rackspace Spot
Rackspace's auction-based cloud where users bid per node-hour for VMs and managed Kubernetes capacity, including GPU server classes with NVIDIA H100 and A30…
Decentralized GPUOnline
Fluence
Fluence Cloud lets teams deploy on-demand GPU and CPU instances sourced from multiple providers via console or API, with zero egress fees and custom GPU…
GPU marketplacesOnline
Lium
GPU rental marketplace where providers list machines that the platform checks and scores, and renters launch GPU pods from RTX 5090 to B300 billed per second…
Decentralized GPUOnline
Nosana
Open-source distributed GPU cloud on which users rent GPUs from independent hosts on demand for AI inference, training, fine-tuning and rendering, deployed via…
- Lowest on-demand H100 rate
- 150 US cents/GPU-h
- Lowest on-demand A100 rate
- 61 US cents/GPU-h
GPU marketplacesOnline
NVIDIA Brev
NVIDIA platform giving access to GPU virtual machines preconfigured with CUDA, Python, Docker and JupyterLab across multiple cloud providers, plus shareable…
GPU neocloudsOnline
Runpod
GPU cloud offering on-demand GPU pods across 30+ regions, serverless GPU endpoints for API inference, and multi-node GPU clusters, with Community Cloud and…
- Lowest on-demand H100 rate
- 199 US cents/GPU-h
- Lowest on-demand A100 rate
- 119 US cents/GPU-h
GPU marketplacesOnline
Spheron
GPU cloud marketplace aggregating capacity from multiple data center providers into one dashboard, offering on-demand, spot and reserved GPU instances plus…
- Lowest on-demand H100 rate
- 420 US cents/GPU-h
- Lowest on-demand A100 rate
- 148 US cents/GPU-h
GPU marketplacesOnline
Vast.ai
GPU marketplace where prices are set by supply and demand across 40+ data centers, offering on-demand, interruptible and reserved instances, plus serverless…
Decentralized GPUOnline
Akash Network
Decentralized compute marketplace where independent providers bid to host containerized workloads on GPUs and CPUs; also offers AkashML, OpenAI-compatible…
Decentralized GPUOnline
Flux (FluxEdge)
Flux is a decentralized cloud of user-operated nodes; its FluxEdge product is a P2P compute marketplace connecting developers with hardware providers renting…
GPU marketplacesOnline
GPUNET
GPU aggregator bringing bare-metal GPUs, datacenters and colocation from multiple partners under one contract, with on-demand GPU rental and reserved clusters…
- Lowest on-demand H100 rate
- 220 US cents/GPU-h
- Lowest on-demand A100 rate
- 162 US cents/GPU-h
Bare-metal GPU clustersOnline
Hydra Host
Bare-metal GPU platform (Brokkr) aggregating capacity from independent data center operators across 40+ data centers, offering NVIDIA and AMD GPUs on…
GPU marketplacesOnline
Hyperbolic
AI compute platform offering on-demand and reserved NVIDIA H100, H200 and B200 GPUs sourced from a supplier network, managed Kubernetes with GPU worker nodes,…
- Lowest on-demand H100 rate
- 319 US cents/GPU-h
Decentralized GPUOnline
io.net
Decentralized GPU cloud aggregating GPUs from many suppliers; offers on-demand GPU clusters (Ray, Kubernetes, containers, bare metal) plus io.intelligence…
- Lowest on-demand H100 rate
- 219 US cents/GPU-h
Decentralized GPUOnline
OctaSpace
Decentralized GPU cloud where users rent GPU nodes from a global network with live hourly pricing for AI, rendering (OctaRender) and app workloads, accessed…
- Lowest on-demand H100 rate
- 12 US cents/GPU-h
- Lowest on-demand A100 rate
- 48 US cents/GPU-h
GPU marketplacesOnline
Prime Intellect
AI training, inference and compute platform offering on-demand access to 1-256 GPUs across clouds, reserved clusters quoted from 50+ providers, a spot resale…
GPU marketplacesOnline
Shadeform
GPU cloud marketplace for provisioning GPU instances across 30+ cloud providers and 100+ regions from one console and API with centralized billing; reserved…
GPU marketplacesOnline
TensorDock
Marketplace GPU cloud of vetted independent hosts offering KVM virtual machines with dedicated GPUs and root access across 100+ locations in over 20 countries,…
- Lowest on-demand H100 rate
- 225 US cents/GPU-h
- Lowest on-demand A100 rate
- 180 US cents/GPU-h
GPU marketplacesOnline
QuickPod
Distributed GPU and CPU marketplace for renting compute from host machines, with templates such as Jupyter for PyTorch and TensorFlow, persistent storage,…
GPU neocloudsOnline
SF Compute
GPU cloud that builds data centers and GPU clusters, offering VMs, bare metal and managed Slurm, with a compute market that lets customers resell reserved…
GPU marketplacesOnline
NVIDIA DGX Cloud Lepton
NVIDIA platform connecting developers to GPU compute across cloud partners and marketplaces, with a unified workflow for AI development, training and inference…
GPU marketplaces by city
Related categories
Questions
Which cities have the most GPU marketplaces?
The cities with the most GPU marketplaces are Online, Global (15 GPU cloud providers), Online, Global (2 GPU cloud providers), Online, Global (1 GPU cloud provider), Online, Global (1 GPU cloud provider), and Online, Global (1 GPU cloud provider).
Published by GPUClouds. How we check each GPU cloud provider.