Definition
A deployment mode where the software and data run on the company's servers, on its premises or in its datacenter, rather than in a provider's cloud.
An on-premise deployment installs the AI platform directly on the company's infrastructure: its physical servers, its datacenter or its private cloud. Prompts, documents and search indexes never pass through a third party, it is the strongest confidentiality guarantee that exists.
Long reserved for large groups for reasons of complexity, on-premise has become widely accessible: a platform like Helix installs in a day on standard hardware, with open-source models running locally and, if desired, controlled access to cloud models for non-sensitive tasks.
All AI models, deployed on your infrastructure, with your documents connected.