Get a Blackwell Ultra B300 GPU today with Civo

NVIDIA's latest Blackwell architecture. Deployed on demand. Ready when you are.

GPU illustration representing graphics processing

The inference GPU built for reasoning AI


Accelerate your most demanding inference, agentic AI, and large-scale generative workloads with dedicated B300 compute.


With 288GB of HBM3e memory, 8 TB/s bandwidth, and 14 petaFLOPS of FP4 throughput, the B300 handles models that previously required whole clusters, on a single GPU. Full ML ecosystem support, transparent pricing, no infrastructure headaches.

Clients

Bright minds build with Civo

B300 GPU Pricing

Built for extreme workloads

ProductPrice
B300 GPU Pricing
from $5.45

Build AI faster

The reasoning GPU, deployed your way

The age of AI reasoning demands more from your infrastructure than ever before. With Civo, you can get B300 clusters up and running fast, so your team can stay focused on innovation. We handle the infrastructure so you can handle the breakthroughs.

01

Create, don't configure

Your team should be building models, not managing infrastructure. Get up and running fast with pre-configured B300 compute or Managed Kubernetes. Pre-installed NVIDIA drivers and software mean you go from provisioned to productive in minutes. Stop setting up. Start building.

02

Power without the price drama

Don't pay for idle machines or surprise invoices. With Civo, every hour of B300 compute is put to work on training, inference, and data processing. We include data ingress and egress as standard, so you can scale without inflating your budget. No drama, just compute.

Get started today

NVIDIA B300 GPUs are in stock and ready for deployment. Contact our sales team to secure yours.

NVIDIA DGX B300 specifications

Take an in-depth look

GPU
8x NVIDIA Blackwell Ultra SXM
CPU
Intel® Xeon® 6776P Processors
Total GPU memory
2.1 TB
Performance
FP4 Tensor Core: 144 PFLOPS | 108 PFLOPS* FP8 Tensor Core: 72 PFLOPS**
NVIDIA NVLink™ switch system
2x
NVIDIA NVLink bandwidth
14.4 TB/s aggregate bandwidth
Networking
8x OSFP ports serving 8x single-port NVIDIA ConnectX-8 VPI > Up to 800 Gb/s NVIDIA InfiniBand/Ethernet 2x dual-port QSFP112 NVIDIA BlueField-3 DPU > Up to 400 Gb/s NVIDIA InfiniBand/Ethernet
Management network
1GbE onboard NIC with RJ45 1GbE RJ45 Host baseboard management controller (BMC)
Storage
OS: 2x 1.9 TB NVMe M.2 Internal storage: 8x 3.84 TB NVMe E1.S
Power consumption
~14 kW
Software
NVIDIA AI Enterprise—Optimized AI software NVIDIA Mission Control—AI data center operations and orchestration with NVIDIA Run:ai technology NVIDIA DGX OS—Operating system Supports Red Hat Enterprise Linux/Rocky/Ubuntu
Rack units
10U
Support
Three-year business-standard hardware and software support

*Specification shown in sparse | dense
**Specification in sparse. Dense is ½ sparse spec shown.

Trusted by top-tier teams worldwide

Serious power for serious workloads

We’re trusted by ambitious startups and major enterprises around the world to deliver scalable, sustainable infrastructure without the hidden costs or headaches of hyperscalers.

Orbital Materials logo
Daniel Miodovnik profile

Daniel Miodovnik

Chief Operating Officer

"We're using Civo's GPUs to develop world-leading AI models to discover new materials and develop hardware solutions to the biggest challenges in data centres."

Resources

Brighter AI starts here

Get in touch today

Let there be cloud

Your faster, fairer, simpler cloud starts here. Talk to a Civo specialist and get started today.

Loading form...

FAQs

Frequently asked questions

Got questions? We’ve got answers.

How can I access the NVIDIA B300 GPU?

To get started with the NVIDIA B300, fill out our form with your requirements and use case. Our team will get in touch to discuss next steps and make sure you have the right setup for your AI and ML needs.