CYANIS™ Cloud

One API, one SLA, sovereign by design.

Sovereign AI inference on distributed compute, routed by the proprietary CYANIS™ orchestration platform.

Join the beta waitlist
Routing modes

Choose how every request is routed

One integration, four routing modes. You set the policy; the orchestration platform places the work.

Sovereign

EU-resident capacity only. Data residency enforced by placement policy; compute steered as a grid-supportive load where sites allow it.

Cost-efficient Global

Lowest-cost global routing. Requests matched to the smallest model tier that clears the quality bar.

Green Compute

Compute only goes to verified GPU powered by renewables.

Sovereignty and Green compute are properties of the mode you choose. Requests in Sovereign mode never leave EU-resident capacity; Global mode routes worldwide for lowest cost.

Alpha results

What the platform measured

Above 96%
cross-server scaling efficiency, including a 675B-parameter model
150 ms
or 28%, lower median time to first token from regional placement
19,468
requests with zero errors under deliberate full saturation
Up to 72%
lower compute at the measured high-offload operating point, with about 79.5% benchmark retention

Measured during the July 2026 Alpha on European cloud infrastructure across two independent providers. The compute figure is benchmark-derived from the measured routing distribution and model memory footprints, and is always quoted together with its retention.
Read the full Technical Brief

Pricing

Three ways to buy compute

Transparent, retail token pricing. Final rates are published at launch.

Credits

Transparent per-model token pricing, billed from prepaid credits.

Per-model rates published at launch

Batch

Asynchronous processing at up to 50 percent below realtime pricing.

Up to 50% below realtime

Reserved capacity

Monthly committed volume with guaranteed capacity and a contractual SLA.

Contact us
Quickstart

One endpoint, streaming by default

Point your existing client at the CYANIS™ gateway and pick a routing mode.

chat completions 
curl https://api.cyanis.ai/v1/chat/completions \
  -H "Authorization: Bearer $CYANIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "cyanis-sovereign",
    "stream": true,
    "messages": [
      { "role": "user", "content": "Summarise this contract clause." }
    ]
  }'

Full documentation at launch; the API is OpenAI-compatible.

Beta access

Join the beta waitlist

Tell us how you plan to use CYANIS™ Cloud and we will be in touch as beta capacity opens.

Sovereignty requirement *
Trust

Built for enterprise review

DPA and security summary available on request; deeper technical detail under NDA.

CYANIS™ Cloud runs today on European cloud capacity and increasingly on energy-linked CYANIS™ sites.