What is ZenithAI? An overview

ZenithAI is a private AI platform that runs on servers your organisation owns. Your people use it in the browser to chat, ask questions about documents, search chosen websites, and create Word, Excel and PowerPoint files. Developers use the same features through an HTTPS API. Your data stays apne ghar mein, at home on your own server.

How it fits together

• Your ZenithAI server has its own address, such as https://dummydomain. Type yours in the Your server box in the sidebar and every code sample on this site will use it.
• Tiers: requests are answered by Instant, Fast, Smart or Genius. You pick by name; see Choose a tier.
• Accounts: an administrator creates them. Developers create API keys that act on their behalf.

For teams at work

• Chat, with files, images, audio and video
• Workspaces, your team’s shelf of documents to ask questions about
• Audio and minutes of meeting
• Site-locked chat, answers only from websites you trust
• Web search and internet
• Word documents, Excel workbooks and PowerPoint decks

For developers

Start with the Quickstart. Already using OpenAI libraries? Read Use your existing OpenAI client.

For IT and administrators

Hardware, installation, users and admin and licensing.

Common questions

What is ZenithAI?

ZenithAI is a private AI platform that runs on servers your organisation owns. Your team uses it to chat, ask questions about documents, search chosen websites and create Word, Excel and PowerPoint files. Developers use the same features through an API.

Where is our data kept?

On your own ZenithAI server. Chats, documents and files stay there. Internet search is used only when it is switched on, and administrators can turn it off for any person.

What hardware does ZenithAI need?

At least two NVIDIA RTX PRO 5000 Blackwell graphics cards with 48 GB of memory each, 128 GB of RAM and a 1 TB SSD, running Linux. See the hardware page for the full list.