Sizing, procurement and managed service on one timeline
Nucleus sizes the GPU hardware to your workload, installs it at your premises, and maintains the stack with an SLA behind it. One conversation covers the whole engagement.
One conversation, one timeline
Sizing, hardware, installation, SLA and compliance are covered in a single engagement.
Maintained by Nucleus
Nucleus engineers handle delivery, installation, burn-in, scheduled maintenance and model updates on site.
SLA-backed support
Enterprise agreements carry a 99.9% uptime commitment with service credits and tiered response times.
Built for data that cannot leave
The entire inference path stays on hardware you physically control, fully air-gapped when required.
What the engagement covers
Nucleus covers the full engagement in four steps. Each is a defined deliverable with a clear handover point. Step one is sizing: we profile your workload, model size and throughput requirements, and recommend a hardware configuration. Step two is procurement: hardware is ordered, built and loaded with your fine-tuned model before it ships. Step three is installation: Nucleus engineers deliver, rack, cable and commission the hardware at your site. Step four is handover: your team receives a running OpenAI-compatible endpoint on your LAN, documentation and the maintenance schedule.
Results from LLMs powered by Nucleus
downtime prevented this month
rerouted in a single fleet optimisation
monthly revenue re-attributed in marketing
prediction accuracy on equipment failures
Enterprise workflows, powered by Nucleus
Line 3's next failure, booked into the maintenance window
47 trucks rerouted while they were still moving
At a glance
The engagement
- Sizing, procurement, installation and handover
- Dedicated GPU hardware at your premises
- Managed maintenance and model updates
The endpoint
- OpenAI-compatible API on your LAN
- Fine-tuned model, pre-loaded
- Fully air-gapped when required
SLA and support
- 99.9% uptime commitment
- Service credits from 10% to 30%
- Enterprise response times: 1 hour urgent, 24/7
Compliance
- Data encrypted at rest (AES-256) and in transit (TLS)
- PDPA compliant
- Role-based access control
- No runtime connection back to Nucleus
On-premise pricing
On-premise platform licence
Your GPU and cloud infrastructure are billed separately.
Pick your path
Custom AI Solutions
Full-stack applications on your private model
Industry solutions
Live workflows across industries
Contact sales
Tell us what you want to build
Learn more
Ready to bring the model on site?
Talk to our team about sizing, procurement and the managed service.