Get Started
ConfidentialMind is a self-hosted AI platform for running large language models (LLMs) and AI applications on your own infrastructure: on-premises, air-gapped, or in your own cloud account. It is built for companies and public-sector organizations that want to use AI but cannot send confidential data to external AI services, so sensitive data never leaves your environment.
What does ConfidentialMind offer?
Everything is managed and interacted from a single web portal, where you can:
- Deploy AI models on your own hardware quickly.
- Connect external AI providers and manage them alongside your own models.
- Build AI systems on your own data with preferred models and integrate them into your software through standard HTTP APIs.
- Chat with AI through a ready-to-use web app. Choose the models, tools, and data for each task to get better results, with no coding required.
- Create AI agents for specific tasks, like answering support questions or summarizing reports, each powered by the model you choose.
- Monitor and control usage with dashboards for better resource management and cost efficiency.
You don’t need to be an infrastructure expert. ConfidentialMind handles setup, security, and access control behind the scenes, so your teams can focus on using AI instead of maintaining its complex infrastructure.
Choose your path
Pick the path that matches what you want to do. Each one leads into the pages you need next.
Learn the core concepts
Everyone
Understand tenants, the Model Gateway, RAG endpoints, agents and API keys.
Use the platform
End users
Chat with a model and ask questions over your own data.
Build with the AI APIs
Developers
Get an API key and make your first Model Gateway call.
Manage the tenant
Tenant administrators
Invite users, organize them into groups, and control who can use each service.
Administer the platform
Platform administrators
Deploy models, manage model catalog, and monitor the usage.
Install and operate
Infrastructure teams
Set up the platform on your cluster.