Are your prompts encrypted?
Attachment
Attach a human.
Leave an email and a person replies.
Sent to Valarian under its privacy policy.
Encryption ends where inference begins.Inference and output inside your perimeter.
The prompt leaves your VPC encrypted. The model reads it in English, outside your environment.There is no external inference endpoint. You own the gateway, the compute, and the policy.
Your VPC
Applications / agents
S3 / data stores
PrivateLinkYour account
The wire
Encrypted · TLS
Managed inferenceDedicated GPUs
Bedrock
AI Foundry
Vertex AI
Welcome to the frontier of enterprise AI.
ACRA makes it possible to scale frontier intelligence inside an environment you control, without an external inference endpoint.
01
Run on dedicated GPUs.
LLMs don’t work with encrypted prompts. If your model is served outside your environment, your model provider can see your unencrypted data.
ACRA enables you to route and schedule your workloads across NVIDIA, Cerebras, and more — all inside your environment.
02
Choose your model.
Download any open-weight model on Hugging Face and start running it inside your environment. Pick models suited for your work, fine-tune them inside your boundary, and swap between them freely without losing context.
You can still use frontier models like Fable and Astra inside ACRA if you choose. They serve outside your environment, but you control what they have access to, every call is recorded, and you can cut the route at any time.
03
Connect your tools.
Turn intelligence into outcomes. Bring the harness you already use, with its loop, tools, skills, memory, and prompts, and connect it to your context: MCP servers, APIs, data stores, other agents. Declare each connection for that workload alone; ACRA grants it, and you can revoke it in one step.
04
Scale on Kubernetes.
ACRA extends the Kubernetes you already run, and lets you have hundreds of concurrent users, several models, GPU fleets scheduled and shared, failover and audit, and the same experience for everyone. As the number of workloads grows, the control architecture does not.
ACRA is your control layer.
Control without scale is a bunker. Scale without control is a dependency. ACRA is both.
| Capability | Frontier APIAPIAnthropic, OpenAI, xAI | Managed inferenceManagedBedrock, AI Foundry, Vertex AI | Locally hostedLocalopen weights, self-managed | Private modelPrivateCohere, Mistral, Aleph Alpha | ACRAACRAsovereign execution |
|---|---|---|---|---|---|
| Access frontier models | Yes | Yes | No | No | Yes |
| Run any open-weight model | No | Yes | Yes | Yes | Yes |
| No external inference endpoints | No | No | Yes | Yes | Yes |
| Your data can’t be used as training data | No | No | Yes | Yes | Yes |
| Policy enforced by the environment | No | No | No | No | Yes |
| Governs agents and tools | No | No | No | No | Yes |
| Deploy at organisational scale | Yes | Yes | No | Yes | Yes |
† Prevented by the provider’s terms as published in September 2026, not by your architecture. A lab’s model through Bedrock, AI Foundry, Vertex AI, or its own API, sits in the first two columns.
