
Graphcore
Best for Researchers and teams running highly sparse machine learning models that struggle on GPUs.

Best for Developers seeking predictable pricing and ultra-fast storage for ML tasks.
UpCloud is a high-performance cloud provider recognized for its proprietary MaxIOPS block storage technology. Catering primarily to developers and SMBs, UpCloud has aggressively expanded its GPU offerings in Europe and globally. Known for extreme reliability, exceptional customer support, and transparent pricing with zero hidden egress fees, UpCloud provides a streamlined, cost-effective alternative to AWS or GCP for deploying AI inference nodes and scaling web applications.
| GPU Models | L40S, A100 |
| GPU Types | A100, L40S |
| Headquarters | Helsinki, Finland |
| Founded | 2011 |
| Availability | Available Now |
| Website | upcloud.com ↗ |
💡 Pricing note: Rates shown are indicative. Final pricing depends on GPU model, reservation type (spot vs. on-demand), contract length, and region. Get an exact quote →
UpCloud GPU cloud pricing starts from $1.20/hr depending on GPU type, reservation model (on-demand vs. spot vs. reserved), and region. Use the quote form to get exact pricing for your specific workload.
UpCloud offers L40S, A100 GPU instances. Availability varies by region and configuration. Contact the provider through ComputeStacker for current availability.
UpCloud operates data centers in Asia, EU Central, US West. Choosing a region close to your users minimises latency and can help with data residency compliance requirements.
Use the "Get a Quote" button on this page to submit your GPU requirements. ComputeStacker will forward your request to UpCloud and other matching providers. You'll receive proposals within 24 hours — no commitment required.
UpCloud offers high-performance GPU infrastructure suitable for large language model training and fine-tuning workloads. For large-scale distributed training, check the Specs tab for NVLink and InfiniBand interconnect availability.

Best for Researchers and teams running highly sparse machine learning models that struggle on GPUs.

Best for AI engineers and studios requiring raw, un-virtualized bare-metal access to the latest NVIDIA H100 and Ada architecture.

Best for Web3 AI engineers looking for trustless, decentralized training networks.