🏭 Large scale · Total sovereignty

MIIA ENTERPRISE

Sovereign AI infrastructure for large organizations, corporations, and governments

Private nodes and clusters. No dependency on any cloud provider. Full compliance.

The listed prices are reference configurations. Each enterprise deployment is customized based on workload, availability, regulatory compliance, and sector requirements. Contact our engineering team for a tailored proposal.

Sectors where MIIA ENTERPRISE operates

🏛️ Public administration

Digital sovereignty, ENS, European regulatory compliance

🏦 Banking and finance

Risk, compliance, anti-fraud, sensitive data models

🏥 Healthcare and research

Clinical data, research, privacy by design

Energy and utilities

SCADA, predictive maintenance, network optimization

🛡️ Defense and security

Air-gap, classified networks, NIS2/DORA compliance

🏭 Industry and manufacturing

Production line AI, computer vision, QA

Reference configurations

Starting point for custom proposals. Each project is unique.

MIIA ENTERPRISE NODE

High-density enterprise inference node. Reference configuration

desde 24.000 €
Processor 2x AMD EPYC 9654 (192 cores)
RAM memory 512 GB DDR5 ECC Registered
Storage 16 TB NVMe U.2 RAID-10
GPU 4x NVIDIA A100 80 GB (320 GB VRAM)
AI models Llama 70B Q8 + DeepSeek V3
Power consumption 3-5 kW (con GPU)
Users 500+ concurrentes
Format Rack 4U · CPD / colocation

Opciones ampliables

GPU options2 opciónes
  • 4x NVIDIA H100 80 GB SXM+75.000 €
  • 8x NVIDIA A100 40 GB+45.000 €
Enterprise networking2 opciónes
  • InfiniBand HDR 200 Gbps+5500 €
  • 25 GbE dual-port NIC+1500 €
Enterprise services7 opciónes
  • Enterprise admin panel (multi-tenant)included
  • API gateway + rate limitingincluded
  • 99.9% SLA · 24/7 support 2hincluded
  • SSO / SAML / LDAP integrationincluded
  • Fine-tuning + custom training+8000 €
  • ISO 27001 / ENS compliance+4500 €
  • Annual security audit+3200 €

MIIA ENTERPRISE CLUSTER

Multi-node cluster with distributed orchestration. Unlimited horizontal scaling

desde 95.000 €
Base nodes 4x nodos NODE (ampliable)
Total RAM 2 TB DDR5 ECC (base)
Storage Ceph distribuido 100 TB base
Total VRAM 1.28 TB VRAM GPU (base 16x A100)
Models Multi-model orquestado (vLLM)
Power consumption 12-25 kW (base)
Users Miles de concurrentes
Orchestration Kubernetes / Slurm / Ray

Opciones ampliables

Cluster scale3 opciónes
  • 8 nodes (x2 base)+115.000 €
  • 16 nodes (x4 base)+340.000 €
  • Additional H100 x8 GPU node+230.000 €
Enterprise cluster services6 opciónes
  • Kubernetes + MIIA Operator orchestrationincluded
  • MLOps pipeline (MLflow + ArgoCD)included
  • Grafana + Prometheus monitoringincluded
  • On-site implementation project+12.000 €
  • Annual maintenance contract+18.000 €
  • Technical team training (10 days)+8500 €
0
Data sent to external cloud
100%
Control over your models and data
99.9%
Guaranteed SLA uptime
24/7
Technical support included

Ready to deploy sovereign AI?

Our engineering team analyzes your requirements and designs the ideal cluster for your organization.

Frequently asked questions

Can I deploy on-premises or in colocation?

Yes. We offer deployment in your data centers, third-party colocation, or air-gapped environments without external connectivity.

What SLA do you offer?

99.9% uptime SLA, 2h response time for critical incidents, 24/7 support. We can negotiate stricter SLAs depending on the contract.

Does it comply with ENS, ISO 27001 or NIS2?

We design deployments to comply with required regulatory frameworks. We offer compliance audits and documentation.

Is it possible to train custom models?

Yes. We offer supervised fine-tuning, RLHF, and full training on your hardware with your proprietary data.

How long does the implementation project take?

Depends on complexity. A simple node: 2-4 weeks. A distributed cluster: 2-4 months. Includes planning, deployment, and training.