Answers to the questions we hear most — about data, models, cost and control.
You choose. With the managed cloud we host everything; in hybrid mode your application and data stay with you and only the prompt leaves for inference; on-premise, nothing crosses your network boundary.
All leading models through one interface — Anthropic, OpenAI, Google (Gemini), xAI, Mistral, Meta, DeepSeek and Qwen — using a managed key, your own key (BYOK) or your own LLM endpoint.
Bring Your Own Key: use your own provider keys or a self-hosted endpoint. You keep your negotiated rates and gain one place to govern all usage — only the prompt goes out, never your data.
Set monthly budgets for the company, teams and individuals — enforced automatically before every call. You get full observability of who used which model, how many tokens and at what cost, with CSV export.
Guardrails detect and anonymize personal data (PII), block prompt-injection and jailbreak attempts, and apply compliance templates — on input and output, by default.
Yes. Connect Google Drive, SharePoint, OneDrive and Confluence (plus public data sources); your documents become part of the answer via retrieval (RAG), grounded in your company knowledge.
Roles (Admin, Advanced, Basic), teams and projects give you clear, multi-tenant access control — who may do what, within your rules.
The managed cloud is the fastest way to start; hybrid and on-premise deployments are available when you need more sovereignty. Book a demo and we'll tailor it to your setup.