Bare-metal servers.
Dedicated nodes with zero virtualization overhead. The full silicon — CPU, memory, GPUs, and fabric — belongs to your workload alone. Provisioned in minutes via API, CLI, or console.
Full control, none of the plumbing.
Single-tenant isolation
No noisy neighbors, no shared caches, no co-tenancy surprises. Your node is yours, physically and logically.
Your OS, your stack
Install anything: custom kernels, InfiniBand drivers, your own container runtime. Full root, full freedom.
Metered power
Per-second billing, spot capacity for burst jobs, and reserved fleets with predictable monthly pricing.
API-first provisioning
Spin nodes up and down from your CI pipeline. A single POST request creates a fully networked node.
Fabric networking
Connect nodes over RDMA fabrics with up to 1.2 TB/s per GPU for tightly coupled distributed training.
Agreed SLAs
Service-level commitments are stated in your agreement, with proactive parts replacement and a live engineer on every ticket.
Pick your chassis.
| Config | Accelerator | GPUs/node | Memory/GPU | Best for |
|---|---|---|---|---|
| h100-8 | NVIDIA H100 | 8 | 80 GB | Prototyping, CI, LoRA tuning |
| h200-8 | NVIDIA H200 | 8 | 141 GB | Production fine-tuning, inference |
| b200-8 | NVIDIA B200 | 8 | 192 GB | 70B-class training fabrics |
| b300-8 | NVIDIA B300 | 8 | 288 GB | Frontier pretraining, long-context inference |
Side by side, at a glance.
| Spec | NVIDIA H100 | NVIDIA H200 | NVIDIA B200 | NVIDIA B300 |
|---|---|---|---|---|
| Architecture | Hopper | Hopper | Blackwell | Blackwell Ultra |
| Memory | 80 GB HBM3 | 141 GB HBM3e | 192 GB HBM3e | 288 GB HBM3e |
| Memory bandwidth | 3.35 TB/s | 4.8 TB/s | 8 TB/s | 8 TB/s |
| FP16 / BF16 (dense) | 989 TFLOPS | 989 TFLOPS | 2.25 PFLOPS | 3.5 PFLOPS |
| FP8 (dense) | 1,979 TFLOPS | 1,979 TFLOPS | 4.5 PFLOPS | 7 PFLOPS |
| FP4 (dense) | — | — | 9 PFLOPS | 15 PFLOPS |
| NVLink | 900 GB/s | 900 GB/s | 1.8 TB/s | 1.8 TB/s |
| Power (per GPU) | 700 W | 700 W | 1,000 W | 1,200 W |
| Route | H100 → | H200 → | B200 → | B300 → |
Ready to push the frontier?
Provision bare-metal GPUs in minutes, scale capacity on demand, and get a real engineer on call — not a ticket queue.