Flexible access to NVIDIA® GB300 NVL72
From a single tray to multiple racks
Turnkey infrastructure · Expert support · Powered by NVIDIA®
Configurations
Suitable for ad-hoc experiments and production workloads
-
1x-18x trays
Each tray contains 4x GB300 GPUs and 2x Grace CPUs.
-
1x NVL72 rack
Each rack contains 18x trays with NVLink v5 peer-to-peer connect.
-
+2x NVL72 racks
Multiple racks each containing 72x GPUs with InfiniBand interconnect.
Specifications
Per 1x tray
- 1152 GB GPU VRAM
- 900 GB GPU RAM
- 128 cores 2x Grace CPU
- 1800 gbit NVLink v5
The GB300 Grace-Blackwell superchip features two Blackwell Ultra B300 GPUs and one Grace CPU, an enhancement from the previous generation's configuration of a single Hopper GPU and one Grace CPU. These components are interconnected via NVLink v5, facilitating a unified memory domain. Each GPU is equipped with 288GB of HBM3e memory.
The NVL72 rack integrates 18 trays, each containing two GB300 Grace-Blackwell superchips, totaling four B300 GPUs and two Grace CPUs. Unlike an InfiniBand cluster, the GB300 utilizes NVLink v5 for GPU interconnectivity. This provides significantly higher throughput and lower latency for GPU-to-GPU communication, optimizing performance for demanding computational tasks.
Pricing
Long-term discounts available
Performance
Built for the age of AI reasoning
| Metric | CPU |
|---|---|
| FP4 | 1080 PFLOPS (1.08 EFLOPS) |
| FP8 | 360 PFLOPS |
| FP16/BF16 | 180 PFLOPS |
| FP32 | 3 PFLOPS |
| GPU memory | 20 TB |
| GPU memory bandwidth | 576 TB/s |
| NVLink bandwidth | 130 TB/s |
| Max GPU TDP | 100.8 kW |
GB300 delivers the cutting-edge performance, with substantial gains over the previous generations.
Get the architecture overview and performance implications from our technical blog that covers:
- Per-GPU specifications for Blackwell and Blackwell Ultra GPUs
- Key differences between GB300 and B200
- NVL72 rack-aggregate performance
- NVLink and NCCL test results
Success stories
Battle-tested with open-source projects and select customers
-
Verda's GB300, combined with their infrastructure support, has provided me with an extremely stable experience. We used their cluster while developing DeepSeek v32 RL, and it was consistently reliable. It allowed me to fully focus on development without having to worry about machine setup or infrastructure issues. The experience was truly exceptional.
Yueming Yuan RL Core Developer
-
vLLM on GB300 provided by Verda is amazing! Trillion-param-level open-source monsters like Kimi K2.5 and GLM-5 are within reach easily. Models that used to need a whole rack now scream on one node. The future isn't coming, it's already shipping tokens.
Kaichao You Core Maintainer
Expert support
Offering deep expertise with the Blackwell Ultra Architecture
Verda is among the first NVIDIA GB300 NVL72 providers, offering deep expertise with the Blackwell Ultra architecture. Building upon early adoption of HGX™ B300, we deployed one of the very first GB300 NVL72 systems in the European Union.
We host and operate GB300 NVL72 systems at our data center locations in in the Nordics, powered by 100% renewable energy sources. Our in-house engineering teams handle the full lifecycle, including:
- Hardware installation
- Infrastructure provisioning
- Systems engineering
Verda's first GB300 NVL72 system was battle-tested with production workloads from open-source projects, such as SGLang and vLLM, and select customers. All early users noted stability and support.
With Verda, you can gain reliable and flexible access to GB300 NVL72 systems with expert support and sensible SLAs.
Verda Cloud Platform
Full-stack AI cloud, rethought from scratch
-
Full-stack AI
Flexible architecture for efficient experimentation, training, and inference at any scale. -
Efficient
Cutting-edge hardware with compute, storage, and networking optimized for peak efficiency. -
Developer-first
Web console, developer docs, API, native SDK, Terraform, and more. -
Reliable
Historical uptime of over 99.9% with fair compensation for service disruptions. -
Expert support
Proactive support from our experienced team of ML craftsmen and infrastructure experts. -
AI R&D
In-house expertise from contributing to frontier research and open-source projects. -
Cost-effective
Streamlined GPU access at up to 90% lower costs than hyperscalers. Long-term discounts available. -
Secure and sovereign
European service that complies with GDPR and adheres to ISO 27001, 27017, 27018, and 27701. -
Sustainable
Hosted in efficient Nordic data centers that utilize 100% renewable energy sources.
Verda Stack
Peak efficiency across software, compute, storage, and networking
AI developer ecosystem
Interfaces for provisioning and managing platform resources.
Managed services
Platform layer for running and scaling AI workloads.
Cloud compute
Virtualized resources for user-managed environments.
Core infrastructure
Compute, storage, and networking powering platform operations.
Data center foundation
End-to-end control for predictable cost, performance, and reliability.
In-house AI R&D
How Verda's AI engineers and GPU infrastructure power frontier AI research
- 1X World Model Verda collaborates with 1X on building multi-GPU inference for the 1XWM generative video model.As video quality leads to task success, it enables solving more complex tasks in household autonomy.
- The SGLang Project Verda sponsors SGLang and its collaborators with access to compute resources and infrastructure support.SGLang’s recent explorations into RL training with FP8 and INT4, utilize NVIDIA Hopper, Blackwell, and Blackwell Ultra platforms.
- 1X World Model Challenge Consisting of Verda's in-house engineers, the Revontuli team won both tracks: sampling and compression.For this challenge, the team utilized Verda's Instant Clusters with Blackwell GPUs and InfiniBand interconnect.