.seers-cmp-cookie-data-hol .seers-cmp-banner-bar .seers-cmp-policy-desc a { color: #b94b01 !important; }

Accelerate AI on Kubernetes with Spectro Cloud and NVIDIA

From edge to data center, we put the power of AI at your command.

AI on Kubernetes with NVIDIA and Spectro Cloud
AI infrastructure in minutes with NVIDIA hardware and Palette's K8s platform

AI infrastructure, simplified

Run the world’s most powerful AI hardware on the industry’s most flexible Kubernetes platform. From Jetson™ edge devices to multi‑cloud GPU factories, Palette gives you an effortless path to production.

Book a live demo and see Palette managing NVIDIA stacks in just 30 minutes.

A deep technology partnership you can trust

As an NVIDIA Preferred Partner, Spectro Cloud works hand‑in‑hand with NVIDIA engineering so you can focus on innovation, not integration. Spectro Cloud is an NVIDIA Jetson Ecosystem member and graduate of the NVIDIA Inception program.

Our teams collaborate on performance tuning, reference architectures and customer projects — so you benefit from proven, validated solutions.

NVIDIA Preferred Partner badge
Quote bg pattern
“The Palette platform enables enterprises and telcos to deploy and operate AI infrastructure with greater speed and efficiency.”
Dror  Goldenberg
Senior Vice President of Networking Software
54
%
of enterprises already run AI/ML workloads on Kubernetes
85
%
of global AI compute is powered by NVIDIA’s CUDA ecosystem
$
202
billion
USD spend on AI servers in 2025 according to Gartner forecasts

Kubernetes gives your data‑science teams the agility they love, while operators keep control over cost, performance and security — especially when GPUs are involved.

Tackling the use cases you care about, together

Edge AI with NVIDIA Jetson™

Tiny yet mighty: Jetson boards and modules bring GPU acceleration to cameras, kiosks and robots. Palette EdgeAI packs let you:

  • Spin up full K3s clusters on Jetson Xavier™/Orin™ in minutes
  • Apply over‑the‑air updates safely — even on cellular links
  • Centrally observe thousands of devices alongside your data‑center clusters
Explore Jetson on Palette
Edge AI with NVIDIA Jetson

GPU‑as‑a‑Service: Build your own NVIDIA GPU

Need elastic GPU capacity without vendor lock‑in? Palette’s Cluster Profiles automate NVIDIA driver installation, the GPU Operator, monitoring and quota management on your chosen infrastructure — bare‑metal, VMware, EKS, AKS, or OCI. Provision a new AI workspace in just minutes, then tear it down when training is done.

Key benefits

  • Up to 70 % higher GPU utilization through shared pools
  • Day‑2 operations baked in: OS patching, firmware upgrades, security scans
  • Self‑service via Palette console, Terraform or API
How to build your own NVIDIA GPU

Accelerate with NVIDIA® BlueField® DPUs

NVIDIA BlueField technology unlocks high‑performance networking, storage and security offloads — but only if you can operationalize it. Palette does the heavy lifting for you:

  • Auto‑detect and configure BlueField hardware the moment a node joins a cluster — no manual flashing or BIOS tweaks.
  • One‑click DOCA runtime install via a Palette pack — validated drivers, firmware and libraries delivered in minutes.
  • Unified lifecycle management with GitOps workflows for firmware, drivers and DOCA components across fleets, with zero‑downtime rolling updates and policy‑based gates.

Want the full story? Read our press release, then dive deeper with the blogs How Palette accelerates Kubernetes clusters with NVIDIA BlueField‑3 DPUs and From pixels to packets: the DPU era.

NVIDIA BlueField DPUS

NVIDIA GPU Operator made easy

Stop stitching YAML. Palette ships a GPU Operator Pack with validated drivers, device‑plugin and monitoring stack. Just add the pack to your profile and every worker node gets GPU‑ready automatically — whether you run bare‑metal or clusters‑in‑VMs with KubeVirt.

Read the deep‑dive blog
NVIDIA GPU operator

The first choice for edge computing

Enterprises, startups and government agencies are choosing Spectro Cloud Palette Edge to help them deliver their edge projects.

Dentsply Sirona uses Palette and NVIDIA hardware for dental imaging inference at the edge.

Read the story

RapidAI uses AI for stroke detection in hospitals worldwide.

Watch the video

Access the broader hardware ecosystem

We partner with leading OEMs so you can buy Palette‑ready servers that ship with NVIDIA hardware installed. Talk to us about integrations with your preferred OEM.

Check out our Edge‑in‑a‑Box solution with HPE. Ruggedized Gen‑11 ProLiant with NVIDIA L4 GPU, pre‑loaded Palette Edge cluster.

Hardware partner ecosystem

Recognized as a leader in AI

CRN AI 100 2025 award
CRN AI 100 Winner 
2024 & 2025
GigaOm Radar Leader – Edge Kubernetes -
2024 and 2025
GigaOm Radar Leader – Edge Kubernetes 
2024 & 2025
GigaOm Radar Leader – Managed Kubernetes -
2024 and 2025
GigaOm Radar Leader – Managed Kubernetes 
2024 & 2025
CNCF Certified Kubernetes Distribution & KCSP
CNCF Certified Kubernetes Distribution & KCSP
NVIDIA Preferred Partner 
since 2023
NVIDIA Preferred Partner 
since 2023

Ready to get started?

Book a demo to see how Spectro Cloud can accelerate your NVIDIA AI workloads on Kubernetes — on‑prem, in the cloud or at the edge.



Book a 1:1 demo with an expert