02 Compute
Run AI whereyour rules say.
Start on the connected cloud with customer data in the EU region. Dedicated, private, hybrid and air-gapped placements are designed for and scoped with you.

[ 02 ] ComputeConnected. Dedicated. Air-gapped.
01 In this layer
Where AI runs.
- Connected cloud, EU regionAvailable
- Dedicated cloud and GPU capacityDesigned for
- Private, hybrid and air-gappedDesigned for
02 [ 02.A ]
Start connected, in the EU region.
The connected cloud runs in AWS eu-west-1 (Ireland). It is the quickest way to start and to prove value.
03 [ 02.B ]
Dedicated capacity when the workload needs it.
Dedicated environments and GPU capacity for one organisation, scoped with you. Not self-serve.
04 [ 02.C ]
No outbound link, if that is the rule.
Air-gapped placement with updates on signed media. The Enterprise Intelligence appliance already has an air-gapped install route.
05 See it work
Compute, result by result.
Ask Compute
Run the claims model in the EU region, on dedicated GPUs, with no outbound link.
Illustration
Illustration
Illustration[ 05.A ] Illustrations, not product screens.
What the infrastructure layer includes
- Compute
- GPUs
- Cloud
- Private cloud
- On-premise
- Hybrid
- Networking
- Storage
- Inference
- Security
- Identity
- Deployment
- Observability
Choose where AI runs
Deployment is a decision you make per workload, not a constraint you inherit. The platform is built so that the same models, agents and workflows can move between these options.
- CloudManaged capacity for teams that want to start quickly and prove value first.
- Private cloudAn isolated environment in a public cloud account or a bare-metal footprint, dedicated to your organisation.Read more
- On-premiseInference and agents inside your own data centre for workloads that must not leave it.
- HybridSensitive workloads on infrastructure you control, burst or experimental workloads elsewhere, under one set of policies.
Controls that travel with the workload
Placement is only useful if the controls move with it.
- Identity firstEvery workload, model endpoint and agent runs as a known identity, so access can be granted and revoked.
- Isolation by designNetwork, storage and compute boundaries are set per environment, so one team’s data and models stay out of another’s.
- Inference you can placeServe open and private models on GPUs you choose, close to the data they work on.Read more
- Observability throughoutLogs, traces and cost are captured at the infrastructure layer and feed the audit trail above it.
Dependencies are part of sovereignty
Where a server sits is only part of the picture. Sovereign infrastructure also means knowing which providers, regions and components your AI depends on, and being able to change them. That is why this layer connects to supply-chain sovereignty as well as infrastructure sovereignty.
Where Swfte products sit in this layer
Product names are kept. Each is an entry point into the platform.
- Dedicated cloudIsolated VPC or bare-metal deployment in a public cloud or your own data centre.Open Dedicated cloud
- GPUReference for the GPUs used to serve and train large language models.Open GPU
- Connect: self-deployRun the model gateway in your own cloud, VPC or data centre, designed for on request.Open Connect: self-deploy
What this means for your team
- CTODeploy models, agents and workflows on infrastructure you control.
- CISOGive AI identity, permissions, policies and auditable controls.
- CIOBuild one controlled environment for enterprise AI.
Frequently asked questions
What does sovereign AI infrastructure mean?
It means retaining meaningful control over where AI runs and what it depends on: compute, hosting, deployment and critical dependencies. It is about control, not only about the location of a server.
Can AI run on-premise or in a private cloud?
The platform is designed for cloud, private cloud, on-premise and hybrid deployment. The right option depends on your workload, jurisdiction and security policy, and we scope it with you.
Does Swfte offer a private GPU cloud?
Private and dedicated GPU capacity for inference is part of the platform’s position. Talk to our team about what is available for your region and workload.
Where is my data stored?
It depends on your deployment. For what is true today about hosting regions and what is still in progress, see the trust centre. Data residency options are part of the platform’s design, and are agreed per deployment.
How does infrastructure connect to the layers above it?
Placement, isolation and identity set here apply to data, models, agents and workflows. Governance runs through all of them, so a policy decision made above is designed to be enforced on the infrastructure below.
Across every layer
These ideas apply to every layer of the platform.
- Controlled autonomyFive levels, from AI that recommends to AI that adapts within limits.Open
- Trust ProfileThe record that describes what each AI system is and may do.Open
- AI sovereigntySeven kinds of control over your AI estate.Open
- AI governanceGovernance that runs inside AI, not beside it.Open
- Swfte Intelligence PlatformAnalyse and visualise your data, usage, agents and outcomes, then build from what you see.Open
- Company brainOne governed place for everything your organisation knows, for every product to read from.Open
- Custom modelsModels adapted on your own data, evaluated before release and deployed under your control.Open
- How Swfte builds with CortexAn honest first-party account of how we build, review and approve work, labelled by status.Open
- Governed agentsIdentity, permissions, policy, approvals and audit for agents, with ten templates.Open
- Compliance-approved workflowsWorkflows with approval gates and evidence built in, with twelve templates.Open

[ 06 ] One platform
Capability and control, in one platform.
Start with one layer and prove it. Add the next when you are ready. The controls come with you.