Skip to main content

What is CaseDesk

CaseDesk gives your organisation a private AI inference endpoint powered by open-source models. Your data never leaves your chosen region.

Deployment modes​

Managed CaseDesk — CaseDesk provisions and runs a dedicated cluster on AWS in your chosen region. You get an OpenAI-compatible endpoint within minutes. Flat monthly subscription — no per-token or per-GPU-hour charges.

CaseDesk Enterprise — For NHS Trusts, government, and regulated industries. Your cloud account, full data sovereignty. Sold through a consultative process — not self-service.

Connect Your Server — Connect an existing GPU VM or on-prem server. CaseDesk installs the runtime and handles endpoint routing. Inference is free once your server is connected — you pay only your cloud provider.

Key capabilities​

  • OpenAI-compatible API — drop-in replacement for OpenAI endpoints in any application
  • Production API key — cd_live_... format key per deployment, shown on the deployment detail page and regeneratable at any time
  • Routing pool — deploy multiple models behind a single endpoint URL; the request's model field selects which model handles it
  • vLLM continuous batching — Team and Advanced tiers serve up to 20 or 50 concurrent users from a single pod, with no per-seat charges
  • Scale to zero — managed deployments scale down when idle and wake automatically on the next request
  • Data residency — UK (AWS eu-west-2, DSPT compliant), EU (Azure westeurope), or US (GCP us-east1)

How CaseDesk compares​

Evaluating hosted alternatives? See how CaseDesk stacks up: