Logiciel Contact Us
Success Stories Tech News Contact Us
whitepaper

Self-Hosting Open Weights Transfers Responsibility, It Does Not Add Security.

Open weights now sit 1.7% behind the best closed model, which makes self-hosting a real option for regulated organisations. This framework prices both paths on measured traffic, names the four drivers that survive scrutiny, and sets out where the hybrid line belongs in your estate.

In depth

The Capability Argument Closed. The Operating Argument Barely Started.

01

The trap most regulated organisations walk into: set the monthly API bill against hardware amortisation, size the cluster for peak from a pilot week, and call the result sovereignty, then run at 19% utilisation while marginal tokens feel free and the true cost per token sits at three to six times nameplate.

02

What the teams doing better do instead: start from the named clause or accreditation condition that hardware actually discharges, measure four full weeks of traffic before sizing anything, build the gateway before the cluster, and price cost per completed task with idle hours in the denominator.

The detail

What Separates A Sovereignty Case That Holds Up.

Zone · 01

Obligations Rather Than Preferences

A standard processing agreement with region pinning and zero-retention terms already covers a great deal of GDPR exposure. Hardware becomes necessary when something specific bites: national cloud rules naming approved operators, a defence classification level, sector limits on where regulated records may be processed, or a subprocessor list that can change on thirty days notice.

Zone · 02

The Artefact You Validated

Pinned

A hosted version can be retired, deprecated or quietly adjusted while your prompts, thresholds and validation evidence stay where they were. For a marketing summary that is an inconvenience. For a credit decision tool a regulator has reviewed, the artefact you validated no longer exists and a decision from eight months ago cannot be reproduced.

Zone · 03

Serving Treated As Production

It never leaves our network is a data flow statement, not a posture. Host it and you own the serving stack: model servers with a history of deserialisation and remote code execution bugs, weights pulled from public registries, GPU drivers, container runtimes, and a KV cache that leaks across tenants when isolation is careless.

By the numbers

The figures that make it a board-level conversation.

15-30%
typical utilisation on a peak-sized self-hosted cluster
3x
shift towards small task-specific models against general-purpose ones by 2027, and most regulated workloads are narrow rather than general
7%
of global annual turnover is the penalty ceiling, with high-risk logging, documentation and traceability obligations phasing in through 2026 and 2027
Inside the report

What you'll take away.

01

Map the obligation before the architecture

Start from the named clause or accreditation condition, and write down what a contract genuinely cannot discharge. If no obligation survives that test, the case is a preference dressed as architecture and the hosted path is cheaper.

02

Measure four weeks of traffic

Peak-to-mean ratio over four full weeks, not a pilot week, which will oversize the cluster by a multiple. Then price cost per thousand completed tasks with idle hours in the denominator, because that is the number finance compares.

03

Build the gateway before the cluster

One internal API holding the policy: data classification in, model selection out, with per-team quotas and cost ceilings. One evaluation harness and one audit log across both backends, so moving a workload becomes a policy change instead of a rewrite.

04

Publish a model bill of materials

Identifier, version, licence, source registry, artefact hash, quantisation, tokeniser, adapters and serving engine version. Verify the hash at load and log it with every batch, so the audit trail names the exact weights behind any response a regulator asks about.

Questions

Frequently asked.

Is self-hosting not automatically more secure because data stays inside our network?

It moves the risk rather than reducing it. You remove one egress path and take ownership of model servers with remote code execution history, public registry weights, GPU drivers, cache isolation between tenants, and all the input and output filtering the provider ran for you.

Our processing agreement pins the EU region already, so do we still need our own cluster?

Usually not. Region pinning with zero-retention terms discharges much of the GDPR exposure. Hardware becomes necessary when an obligation names approved operators, sets a classification level, or forbids a subprocessor list
that can change on thirty days notice.

At what point does a self-hosted cluster actually become cheaper than an API?

At sustained round-the-clock load. An eight-accelerator node costs near $200,000 a year before serving a token, and peak-sized clusters run at 15% to 30% utilisation, so real cost per token is three to six times nameplate. Measure four weeks of traffic first.

Which workloads should stay on a hosted frontier model even after we self-host?

Long-horizon agentic work, complex multi-step tool use and difficult code generation. The capability gap still decides whether those workflows complete at all. Route them through the same gateway with redaction at the boundary, and keep one eval harness across both backends.

Can we treat an open-weight release like any other open source dependency?

Licence terms are where that assumption breaks. They vary widely, some restrict commercial use above a threshold or derivative models, and most releases do not disclose training data. Record the provenance gap, the compensating controls and who accepted the residual risk in your technical file.

Get the whitepaper

Have it emailed to you.

Drop your details and we'll send Self-Hosting Open Weights Transfers Responsibility, It Does Not Add Security straight to your inbox - no spam, unsubscribe anytime.

Download whitepaper
Next step

Decide on measured load.

Bring your traffic profile and the obligation you are trying to discharge. Our engineers will cost both paths and tell you where the hybrid line sits.

Book an architecture review