Large-Scale Clusters
Dedicated clusters. InfiniBand included.
Multi-node GPU clusters for distributed training. 8 to 8,000 GPUs with 400 Gb/s interconnects. We manage the hardware.
research-cluster
32x H100 Node 1
Node 2
Node 3
Node 4
400 Gb/s InfiniBand
Why teams choose Lyceum clusters
Dedicated hardware
No noisy neighbours. Your workloads run on isolated, dedicated infrastructure.
400 Gb/s InfiniBand NDR
High-bandwidth, low-latency interconnects for efficient distributed training.
Slurm or Kubernetes
Your choice of orchestration. We configure and manage it for you.
EU data centres
Full GDPR compliance. Data residency in European data centres.
Request a quote
Tell us about your requirements and we'll get back to you within 24 hours.