Quick Start Guide
Follow these six steps to go from zero to a fully configured Your Office AI organization, with your team chatting, calling, and collaborating with AI assistance.
You'll have Your Office AI up and running in minutes. No infrastructure to set up and no configuration files to edit — just sign up and start collaborating.
Setup steps
- Create your account
Open qa.app.yoffice.ai in your browser and choose "Get started free". Sign up with your email address and a password, or continue with Google. Sign-up includes a quick consent step for the terms, privacy policy, and data processing.
- Create or join an organization
An organization is your company's top-level account — an isolated tenant that holds your workspaces, members, knowledge, integrations, and settings. Create one and give it a name, or, if your team already has an organization, accept an invite from an admin. When no organization is active you'll see a selection overlay.
- Invite your team
Open Settings → Organizations → Members and invite people by email. Choose a role — admin, editor, or member — before sending. Invitees get an email with a join link that lands on a token page to accept.
- Start your first chat
Open Chat from the navigation rail. Start a "New chat" for a private AI session, or pick a workspace channel to talk to your team. Type your message and press Enter to send; use Cmd/Ctrl+Enter to submit from anywhere in the composer.
- Talk to the AI assistant
In a new AI chat, type your question and the assistant replies inline. Pick the model from the selector at the top of the conversation, switch the input mode between text, voice, and avatar, and pull in context with #-mentions for agents, prompts, knowledge folders, integrations, skills, and memory.
- Connect an integration
Open Integrations from the navigation rail. Connect Gmail, Google Calendar, Google Drive, or Google Tasks with one-click OAuth today — the rest of the 60+ tool catalog is marked "Coming soon" — then mention connected tools in chat or trigger workflows from their events.
Press ⌘ Enter (Mac) or Ctrl + Enter to send a message, and ⌘ / Ctrl + F to search your chats. See the full keyboard shortcuts reference.
Things to explore next
Once your organization is running and your team is on board, here are a few of the most popular features to explore.
Research
Research is the AI's most powerful capability. Open it from the chat tool strip or the Knowledge studio and the assistant plans a multi-step investigation — generating sub-queries, searching the web (via Brave Search) and your organization's knowledge, scoring sources, and synthesizing a cited report you can export to PDF. Learn more about Research →
Live transcription
Calls can be transcribed in real time on the server, producing live captions — with optional on-the-fly translation — that participants can toggle. Saved transcripts are searchable and become RAG context for the assistant, subject to an access policy you control. Learn more about Transcription →
Canvas
Canvas turns a chat into a living document. The AI drafts and edits text or code beside your conversation, with autosave, numbered version snapshots, and sharing — export to PDF or Markdown when you're done. Learn more about Canvas →
What's next?
Common questions
Which AI models can I use?
Your Office AI supports six providers: OpenAI, Anthropic, Google, Groq, and locally hosted Ollama and Ollama Cloud. Pick a model per conversation; administrators decide which are available and set spend limits per organization. See the AI Assistant guide for the full list of current models.
Can I run AI privately?
Yes. Connect a locally running Ollama instance, or use Cloud, so inference stays off third-party servers. Administrators can also supply organization API keys for OpenAI, Anthropic, Google, or Groq to use your own quota.
Is Your Office AI available on mobile?
Yes — native apps run on iOS and Android, alongside macOS, Windows, Linux, and the web. Chat, AI, video calls, knowledge, and workflows are all available, with the layout adapting to each screen size.
If you get stuck at any step, visit our FAQ.