Meridian Compute
GPU infrastructure for ML teams
High-performance compute,
cloud simplicity
Run and scale AI workloads on bare-metal-level GPU instances - from a single GPU to thousand-GPU clusters - or serve models on production inference endpoints. One platform, predictable pricing, no cluster wrangling.
Why Meridian
Supercomputer performance, without the ops team
Bare-metal level performance
No GPU or network virtualization - up to 20% higher system MFU than comparative benchmarks, so you need less infrastructure for the same result.
Resilient from the start
Automated health checks and node lifecycle management mean 50% fewer interruptions per day across your fleet.
Developer first
OpenAI-compatible API, a real CLI, and docs that get you from signup to first request in minutes.
GPU lineup
Accelerated compute, powered by NVIDIA
GB300 NVL72
72 Blackwell Ultra GPUs + 36 Grace CPUs per liquid-cooled rack. Frontier training and reasoning at the highest scale.
HGX B200 · H200 · H100
From Blackwell to proven Hopper - large-scale training, memory-intensive inference, and cost-efficient fine-tuning.
CPU instances
Intel Xeon and AMD EPYC instances for app backends, data pipelines, batch inference, and automation around your GPU workloads.
Getting started
Launch your first GPU instance from the quickstart, or talk to our team about capacity and reserved pricing.