Data sovereignty

Your data never leaves infrastructure you control.

Not a policy commitment. An architectural fact.

Marigold runs open-weight models on infrastructure you control: a laptop, a server, or your own AWS or Azure account. There is no vendor in the inference path to send data to, misconfigure a policy around, or have a breach on your behalf. Once set up, it runs air-gapped -- inference makes no external request at all.

Sending data to a commercial AI API is, structurally, the same decision as sending it to any third-party processor. Self-hosting removes that decision rather than managing it.

01

Regulatory risk

NHS data frameworks, FCA conduct rules, GDPR restrictions on international transfer, legal professional privilege -- binary, externally imposed constraints. If your data can't leave the building, running the model inside the building is the only architecture that satisfies the rule outright.

02

Exposure risk

Not everything sensitive is regulated -- competitive intelligence, unannounced product work, client relationships. A commercial provider's breach becomes your breach the moment their processor held your data. With nothing sent externally, there's no third party left to have the breach.

03

Training reappearance risk

Commercial providers train on submitted data, and demonstrated cases exist of training data reappearing in outputs. Open-weight models are fixed and published before you run them -- there is no feedback loop, because there is no vendor collecting anything to feed back.

Criterion Self-hosted (Marigold) Commercial AI API
Inference location Your hardware, or your own cloud account Provider's infrastructure
Prompt sent to a third party Never -- none in the path Always
Training on your inputs Never -- fixed published weights Varies, opt-out often required
Third-party data processing agreement Not applicable -- no processor exists Available, still a third party
Works fully offline after setup Yes No
Open-weight models Yes No
Shape Database Workers Outputs
Laptop or server The bundled Postgres The local GPU or CPU Local disk
Startup A managed Postgres such as Supabase Rented GPU instances An S3 bucket
Institution AWS RDS or Azure Database for PostgreSQL, in your account GPU instances in your account S3 or Azure Blob Storage, in your account

Applications do not change between these: only the system configuration does. In every shape the models, the prompts and the outputs stay on infrastructure you own or rent directly. Deployment guide →

We are looking for design partners: organisations running Marigold in their own AWS or Azure accounts, who want help with the deployment and a say in what the platform does next. Become a design partner →

Does self-hosting remove the need for a data processing agreement?

There's no third-party processor in the inference path, so there's nothing to sign a DPA with for that part of the pipeline. You remain sole controller and sole processor of your own inference. That's an architectural fact, not a legal guarantee -- it doesn't by itself make a given use of data lawful, which still depends on your legal basis and sector.

Does Marigold train on my data?

No. Marigold runs open-weight models as published -- it doesn't train, fine-tune, or send inference inputs anywhere for that purpose. There's no feedback loop between what you run and what any model learns next, because there's no vendor on the other end collecting it.

What data actually leaves my machine when I use Marigold?

None, once set up. Models download once to local disk. After that, inference runs entirely on your own hardware and the stack can be disconnected from the network entirely -- see running air-gapped in the getting-started guide.

Can Marigold run in our own AWS or Azure account?

Yes. The database can be AWS RDS or Azure Database for PostgreSQL, the workers GPU instances, and outputs S3 or Azure Blob Storage, all in your own account. Applications do not change between a laptop and a cloud account; only the system configuration does. See Deployment.

Is self-hosting alone enough to be GDPR compliant?

It resolves the cross-border transfer and third-party processor questions, which are usually the hardest part. It doesn't by itself make every use of personal data lawful -- that still depends on your legal basis, retention practice, and sector rules. Marigold does not provide legal advice.

For a fuller treatment of these three risk categories in general -- not specific to Marigold -- Bay Information Systems has written on it at greater length: Why Private Inference?

Nothing to send. Nothing to leak.

Open source. Run it on your own hardware or in your own cloud account.