A yield report tells a competitor exactly how you're doing.
Keep it on your hardware.
Private LLM hosting on a dedicated machine in EU-Central, for fabs and semiconductor suppliers running yield report summarization and process documentation without sending proprietary process data through a shared AI product.
Not legal or trade compliance advice. This page describes GPUwerk's infrastructure and contractual commitments. It is not an opinion on your intellectual property strategy or export-control classification of your process technology. Check with your own legal team before deciding what can run where. See also private LLM hosting for manufacturing and private LLM hosting for aerospace and defense.
Why fabs and their suppliers end up asking about this
A yield curve or a process recipe is often the single most sensitive document a fab produces. The same goes for a supplier's process documentation shared under an NDA.
Yield report summarization
Summarizing a batch of wafer yield data or a lot disposition report for an engineering review, or pulling a pattern out of months of run data. A yield report describes exactly how a line is performing, which is not something most fabs want summarized by a public API.
Process documentation drafting and search
Drafting a process change record, searching an internal process document archive for a prior deviation, or cleaning up a recipe note before it's filed. Tolerances and process parameters tuned over years are exactly the kind of content a public AI tool's terms may allow it to retain or review. See the private AI technical documentation guide →
Supplier and equipment data
Tool logs, maintenance records, and supplier specifications shared under NDA carry confidentiality terms separate from your own IP. A dedicated single-tenant machine means that data is processed on hardware no other tenant shares, with no other company's workload anywhere near it.
What's actually in the contract
For your general counsel or IT lead to review directly.
| Question | Answer |
|---|---|
| Where does it run? | EU-Central, on a dedicated single-tenant machine assigned to your company |
| Who operates it? | PRINT IT! SE, a Societas Europaea registered in Prague, Czech Republic. No US parent entity. |
| Who can reach the instance? | Through the instance itself, only holders of your SSH keys; password login is disabled fleet-wide. GPUwerk keeps infrastructure administrator access to the underlying machine, as on any hosted service, and under the DPA does not use it on your content except at your request for support or where a legal obligation requires it. |
| Does GPUwerk read yield or process data? | No. We host the hardware and, under the DPA, do not access, read, copy, index or analyse workload content. |
| Sub-processors for the workload? | None, listed at /legal/sub-processors |
| DPA (GDPR Art. 28)? | Published at /legal/dpa, no charge |
| Data on termination? | Container and workspace volume deleted from the node, then the node is sanitised before reassignment. Filesystem deletion, not a cryptographic erase; export what you need before terminating, since it isn't reversible. |
| Pricing | Spark-1x from $0.79/hour on demand, billed per minute; Spark-2x from $1.79/hour. A customer-requested stop holds the reservation at 75% of the running rate. See /pricing for full detail. |
Questions we get from semiconductor manufacturers
Is it safe to put yield data or process documentation into an LLM?
That depends on where the model runs and who can reach it, not the model itself. On a GPUwerk dedicated instance in EU-Central, the machine is assigned to you alone and reachable only through your SSH keys; GPUwerk does not access, read, copy, index or analyse what runs on it except at your request for support or where a legal obligation requires it. Whether a specific document is safe to process at all, given your own IP and export-control obligations, is a decision for your own legal or trade compliance team.
Does GPUwerk train on our yield or process data?
No. GPUwerk operates infrastructure, not models. Your documents are processed by software you install on your own instance, and under our data processing agreement GPUwerk does not access that content except at your request for support or where legally required. There is nothing for GPUwerk to train on.
Do you sign a DPA covering process and supplier data?
Yes, a standard GDPR Article 28 DPA is published at /legal/dpa at no charge. There are no sub-processors for instance workloads, listed at /legal/sub-processors.
What about export-controlled process technology?
GPUwerk can tell you where the hardware sits and who can reach it: EU-Central, dedicated, no US entity in the chain. Whether a specific process document is export-controlled, and what that means for where it can be processed, is a determination only your own compliance function can make.
Related reading
Private LLM hosting for manufacturing · Private LLM hosting for aerospace and defense · Private AI for technical documentation
Related pages
Keep the yield data and the model on the same floor.
Talk to the people who run the racks, or start with a pilot and a practical rollout plan.
Talk to us Deploy an instance