# Compare GPU neoclouds Check service regions with each provider. 34 GPU cloud providers. Also on the [web page](https://gpuclouds.fyi/us/global/online/category/gpu-neoclouds). ## GPU cloud providers 1. [CoreWeave](https://gpuclouds.fyi/us/global/online/coreweave.md), GPU neoclouds, Online: AI cloud offering NVIDIA GPU instances (GB300/GB200 NVL72, B200, H200, H100, L40S, A100 and more) on demand and spot, plus managed Kubernetes (CKS), SUNK,… 2. [Lambda](https://gpuclouds.fyi/us/global/online/lambda.md), GPU neoclouds, Online: GPU cloud offering on-demand NVIDIA B200, H100, A100, GH200 and other GPU instances, 1-Click Clusters of 16 to 2,000+ GPUs, private cloud, and managed… 3. [Mithril (formerly Foundry / ML Foundry)](https://gpuclouds.fyi/us/global/online/mithril-formerly-foundry-ml-foundry.md), GPU neoclouds, Online: 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… 4. [Oblivus](https://gpuclouds.fyi/us/global/online/oblivus.md), GPU neoclouds, Online: Offers on-demand and reserved NVIDIA GPU instances for AI training, inference and rendering, with per-minute billing, an API and deployment in Montreal and… 5. [Rackspace Spot](https://gpuclouds.fyi/us/global/online/rackspace-spot.md), GPU marketplaces, Online: 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… 6. [Together AI](https://gpuclouds.fyi/us/global/online/together-ai.md), Inference platforms, Online: AI platform offering serverless and dedicated inference APIs for open models, fine-tuning, and self-serve GPU clusters (H100, H200, B200, B300, GB200/GB300)… 7. [Atlantic.Net](https://gpuclouds.fyi/us/global/online/atlantic-net.md), GPU neoclouds, Online: Hosting provider and NVIDIA cloud partner offering on-demand cloud GPU and dedicated GPU servers with NVIDIA H100 NVL, L40S and A2 GPUs, including… 8. [Atlas Cloud](https://gpuclouds.fyi/us/global/online/atlas-cloud.md), Inference platforms, Online: Atlas Cloud offers a single OpenAI-compatible API for 400+ image, video, audio and language models, plus on-demand GPU instances with NVIDIA H100 SXM and H200… 9. [Beam](https://gpuclouds.fyi/us/global/online/beam.md), Serverless GPU, Online: Cloud platform offering serverless GPU endpoints billed by the millisecond, persistent VMs, sandboxes, on-demand bare-metal or VM GPU machines, reserved… 10. [ChronoScale](https://gpuclouds.fyi/us/global/online/chronoscale.md), Bare-metal GPU clusters, Online: AI infrastructure company spun out of Applied Digital's cloud business, providing dedicated GPU environments for training, inference and HPC, contracted by the… 11. [CloudRift](https://gpuclouds.fyi/us/global/online/cloudrift.md), GPU neoclouds, Online: GPU rental platform and sovereign AI infrastructure software offering on-demand and reserved NVIDIA and AMD GPU instances (VMs, containers, bare metal) billed… 12. [Crusoe](https://gpuclouds.fyi/us/global/online/crusoe.md), GPU neoclouds, Online: AI cloud offering NVIDIA and AMD GPU instances (GB300, B300, H200, H100, A100, L40S, MI300X) on spot, on-demand or reserved terms, plus managed Kubernetes,… 13. [DigitalOcean GPU Droplets](https://gpuclouds.fyi/us/global/online/digitalocean-gpu-droplets.md), GPU neoclouds, Online: Cloud provider offering GPU Droplets with NVIDIA (B300, H200, H100, L40S, RTX 6000/4000 Ada) and AMD Instinct (MI355X, MI350X, MI325X, MI300X) GPUs on-demand,… 14. [GMI Cloud](https://gpuclouds.fyi/us/global/online/gmi-cloud.md), GPU neoclouds, Online: GMI Cloud provides dedicated single-tenant NVIDIA GPUs (H100, H200, B200, GB200, GB300) in GMI-operated data centers, on-demand or reserved, priced per… 15. [Latitude.sh](https://gpuclouds.fyi/us/global/online/latitude-sh.md), Bare-metal GPU clusters, Online: Bare-metal cloud platform offering Metal GPU instances and dedicated GPU clusters (NVIDIA HGX B300, RTX PRO 6000, H100) with hourly, monthly or yearly billing,… 16. [Lightning AI](https://gpuclouds.fyi/us/global/online/lightning-ai.md), GPU neoclouds, Online: AI development platform (Lightning Studios) providing cloud GPU and CPU machines, from T4 and L4 to A100, H100, H200 and B200, billed by the second with… 17. [Massed Compute](https://gpuclouds.fyi/us/global/online/massed-compute.md), GPU neoclouds, Online: GPU cloud that owns and operates its infrastructure, renting NVIDIA GPU instances (B300, B200, H200, H100, A100, L40S, RTX PRO 6000 and more) by the hour with… 18. [Novita AI](https://gpuclouds.fyi/us/global/online/novita-ai.md), Inference platforms, Online: AI cloud offering model inference APIs for LLM, image, video and audio, plus GPU instances and serverless GPUs across 14 global regions, with on-demand and… 19. [Paperspace (DigitalOcean)](https://gpuclouds.fyi/us/global/online/paperspace-digitalocean.md), GPU neoclouds, Online: GPU cloud now part of DigitalOcean, offering on-demand NVIDIA GPU machines (H100, A100, A6000, A5000, A4000, V100 and others) billed hourly, plus Gradient… 20. [Phala Cloud (GPU TEE)](https://gpuclouds.fyi/us/global/online/phala-cloud-gpu-tee.md), GPU neoclouds, Online: Confidential AI cloud running containers, private LLM inference and GPU jobs inside hardware TEEs (Intel TDX, NVIDIA confidential computing) with attestation;… 21. [Runpod](https://gpuclouds.fyi/us/global/online/runpod.md), GPU neoclouds, Online: 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… 22. [Saturn Cloud](https://gpuclouds.fyi/us/global/online/saturn-cloud.md), GPU neoclouds, Online: Hosted data science and ML platform providing Jupyter, IDE and job resources on cloud CPUs and GPUs (H100 via Nebius; T4, V100, A10G via AWS), charged per hour… 23. [Thunder Compute](https://gpuclouds.fyi/us/global/online/thunder-compute.md), GPU neoclouds, Online: GPU cloud offering persistent on-demand GPU instances (RTX A6000, L40, A100 80GB, H100 SXM) billed per minute, and short-lived GPU sandboxes for agents billed… 24. [VALDI](https://gpuclouds.fyi/us/global/online/valdi.md), GPU neoclouds, Online: Provides on-demand GPU virtual machines for AI training, inference and other compute workloads, with per-second usage billing and Ubuntu or Windows… 25. [WhiteFiber](https://gpuclouds.fyi/us/global/online/whitefiber.md), GPU neoclouds, Online: AI and HPC infrastructure provider offering NVIDIA B300, GB300 and VR200 GPU clusters as reserved GPU cloud capacity, turn-key private AI clouds and… 26. [QumulusAI](https://gpuclouds.fyi/us/global/online/qumulusai.md), GPU neoclouds, Online: Atlanta-based NVIDIA Cloud Partner that designs and deploys single-tenant GPU environments with committed capacity, offering NVIDIA B300, B200, H200, H100, RTX… 27. [Rescale](https://gpuclouds.fyi/us/global/online/rescale.md), GPU neoclouds, Online: Cloud HPC and AI platform for engineering simulation that provides access to CPU and NVIDIA GPU architectures (including A100 and V100) deployed across global… 28. [TensorWave](https://gpuclouds.fyi/us/global/online/tensorwave.md), GPU neoclouds, Online: AI cloud built on AMD Instinct accelerators (MI455X, MI355X, MI325X, MI300X) offering bare-metal GPU clusters, managed Kubernetes and Slurm, network storage,… 29. [Voltage Park](https://gpuclouds.fyi/us/global/online/voltage-park.md), GPU neoclouds, Online: GPU cloud (now merged with Lightning AI) offering self-serve on-demand NVIDIA HGX H100 GPUs, dedicated reserved H200/Blackwell clusters, bare-metal access and… 30. [Zenlayer (Distributed Inference / GPU compute)](https://gpuclouds.fyi/us/global/online/zenlayer-distributed-inference-gpu-compute.md), Inference platforms, Online: Zenlayer offers GPU compute for distributed AI inference with NVIDIA RTX 4090, H100 and H200, connected over its global private network spanning Asia, the… 31. [Highrise AI (Hut 8)](https://gpuclouds.fyi/us/global/online/highrise-ai-hut-8.md), GPU neoclouds, Online: Hut 8 subsidiary providing GPU pods and dedicated clusters for AI training and inference, with bare-metal access, managed Kubernetes and reserved capacity. 32. [ionstream](https://gpuclouds.fyi/us/global/online/ionstream.md), Bare-metal GPU clusters, Online: GPU-as-a-service provider offering single-tenant bare-metal NVIDIA HGX B200 and B300 servers, provisioned and networked through its Gravity platform, with… 33. [SF Compute](https://gpuclouds.fyi/us/global/online/sf-compute.md), GPU neoclouds, Online: 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… 34. [Targon](https://gpuclouds.fyi/us/global/online/targon.md), GPU neoclouds, Online: On-demand NVIDIA GPU cloud for training and inference, offering virtual machines with up to eight GPUs, Intel TDX confidential VMs, dedicated bare-metal… ## Other categories in Online - [Inference platforms](https://gpuclouds.fyi/us/global/online/category/inference-platforms.md): 31 GPU cloud providers - [Bare-metal GPU clusters](https://gpuclouds.fyi/us/global/online/category/bare-metal-gpu-clusters.md): 20 GPU cloud providers - [GPU marketplaces](https://gpuclouds.fyi/us/global/online/category/gpu-marketplaces.md): 15 GPU cloud providers - [Serverless GPU](https://gpuclouds.fyi/us/global/online/category/serverless-gpu.md): 15 GPU cloud providers - [Decentralized GPU](https://gpuclouds.fyi/us/global/online/category/decentralized-gpu.md): 9 GPU cloud providers - [Hyperscalers](https://gpuclouds.fyi/us/global/online/category/hyperscalers.md): 5 GPU cloud providers - [Sovereign and regional GPU](https://gpuclouds.fyi/us/global/online/category/sovereign-regional-gpu.md): 1 GPU cloud provider