Self-hosted inference for enterprise AI

Deploy all the models your agents need next to your data, compliant and secure by design.

security review. Approved

From: An insurer in Zurich, your customer
About: The AI agents in your product

Which third parties process our data for AI features?

None. The models run on self-hosted SIE Cluster.
Where is our data processed?

In Zurich, inside Switzerland.
Can the AI features run inside our environment?

Yes, in a private VPC or fully air‑gapped.
Does a model provider see or keep our data?

No. Nothing leaves the cluster.

Reviewed by their security team 4/4 Accepted

Third-party models stall enterprise deals in security review

Your customer	Finding	Reviewer note
Insurer in Switzerland FINMA	Data processed outside Switzerland	Client data has to stay in Switzerland. Your model provider's nearest region is in Frankfurt.
Defense contractor CMMC Level 2	Controlled data leaves the enclave	Controlled unclassified information can't go to a model API outside our accredited boundary.
Hospital trust in the UK NHS DSPT	Unassessed data processor	Patient data only goes to processors we have assessed, and your model provider is not one of them.
Provincial agency in Canada Nova Scotia PIIDPA	Data stored outside Canada	Personal information has to stay in Canada, and the hosted API routes through the US.

The Information Anthropic Data Fears Prompt Nvidia, Palantir and Booz Allen to Restrict Model Use
As paranoia rises over whether Anthropic or OpenAI could learn from their customers' intellectual property, large firms involved in sensitive corporate work, such as Palantir Technologies, Nvidia and Booz Allen Hamilton, have started demanding new guarantees or reducing or eliminating use of the AI firms' most advanced models.
14 September 2026
The Verge Microsoft restricts Claude Fable for employees over data retention concerns
Microsoft's legal teams are evaluating Anthropic's new data retention changes.
10 June 2026
POLITICO Europe EU Parliament blocks AI tools over cyber, privacy fears
AI features on EU devices sent data to cloud, prompting the Parliament's IT support to switch them off.
16 February 2026
The Register Geopolitics push European CIOs to think local on cloud
Majority of customers plan to favor domestic providers as sovereignty fears rise
13 November 2025

One SIE Cluster runs every model your AI platform needs

SIE Cluster One endpoint

Select a model or a task to open its page

Safety and PII gate Blocks unsafe prompts and masks personal data

On SIE Qwen3Guard 4B
Semantic cache Answers a repeated question at once

On SIE Qwen3 Embedding 4B
Intent and complexity Picks the skill, and how strong a model it needs

On SIE GLiNER2.5 Decide
LLM Plans and answers with an open model, calling SIE tasks as tools

On SIE Qwen3.8 27B
Every task as a tool Chat Structured output Doc to Markdown OCR Speech to text Image classify Detect Image questions
Verify Checks the answer before it goes out

On SIE Granite Guardian 2B

Deploy SIE in your cloud or in your customer's

Deployment guide

Terraform for four clouds Modules for AWS, GCP, Azure and Alibaba Cloud provision the cluster and its GPU node pools.
Helm into any cluster Install into an existing EKS, GKE, AKS or on-premises cluster.
Air-gapped install Mirror the images and model cache, then install with no internet egress.
Worker pools per tenant Dedicated workers keep one customer's load from slowing another's.

Interview with Roman Stanek, GoodData's founder and CEO Coming soon

Case study

GoodData is building its AI platform on SIE

In the interview

Why GoodData runs its own AI platform instead of calling a model API
Where open models match frontier models, and where they fall short
How SIE serves the models behind GoodData's AI features
What enterprise customers ask before they turn AI features on

gooddata.ai

Get Native Inference, or self-host SIE Cluster for free

Native Inference

A platform license for SIE Cluster. Our engineers work with yours in a shared Slack channel and help answer your customers' security reviews. SIE itself stays Apache-2.0.

Talk to us

SIE Cluster

Easy & scalable deployment in your own cloud.

Terraform to your cloud in minutes
Apache-2.0, same engine as Serverless
GPU workers scale to zero between jobs
Per-tenant pools, no noisy neighbors

Agent prompt

Deploy SIE to our AWS account with the superlinked/sie/aws Terraform module. Docs: superlinked.com/docs/deployment. Serve Qwen/Qwen3.8-27B-FP8 using the public model configurations.

Deploy SIE to our Google Cloud project with the superlinked/sie/google Terraform module. Docs: superlinked.com/docs/deployment. Serve Qwen/Qwen3.8-27B-FP8 using the public model configurations.

Deploy SIE to our Azure subscription with the superlinked/sie/azure Terraform module. Docs: superlinked.com/docs/deployment. Serve Qwen/Qwen3.8-27B-FP8 using the public model configurations.

Deploy SIE to our Alibaba Cloud account with the superlinked/sie/alicloud Terraform module. Docs: superlinked.com/docs/deployment. Serve Qwen/Qwen3.8-27B-FP8 using the public model configurations.

Deploy guide

SIE Local

Run the open-source server on your own NVIDIA GPU.

NVIDIA GPU
One container, no cluster
Qwen3.8 27B, fully offline
Same SDK and IDs, no code changes

docker run --gpus all -p 8080:8080 ghcr.io/superlinked/sie-server:latest-cuda13-sglang-cu130

Quickstart