Simple, transparent pricing for sovereign AI.
Pay-as-you-go token pricing for production inference, or reserve capacity and dedicated infrastructure for high-volume and regulated workloads. All on UK infrastructure, invoiced in GBP, with no hidden costs.
Token pricing · MiniMax M2.7
Input tokens
Output tokens
Compare that to £18–£23 per million output tokens on the leading US GPU providers.
Scale beyond pay-as-you-go.
For high-volume, latency-critical or regulated workloads, we provision capacity to fit – talk to us about the right shape for your stack.
Dedicated Infrastructure
A private, customer-specific RDU environment – isolated capacity tuned to your workload, with no noisy neighbours and predictable performance.
- Isolated, single-tenant RDU capacity
- Custom model routing and configuration
- Priority support and SLA
Reserved Capacity
Commit to throughput up front and lock in volume pricing – guaranteed headroom for sustained, high-volume production inference.
- Committed throughput, volume-based pricing
- Guaranteed capacity for peak load
- Access to all production models
On-prem Deployment
Run RDU hardware inside your own perimeter – full data residency and control for the most sensitive defence, public-sector and regulated workloads.
- RDU hardware in your data centre
- Air-gapped and sovereign by design
- Implementation and managed-service options
Ready to start building?
Point your stack at our endpoint and run on UK-sovereign hardware in minutes – or talk to us about dedicated, reserved or on-prem capacity.