Your system.
Your Omni Factory.

Omni Factory runs on your computer and helps you before, during, and after a conversation. Keep using your usual calling or messaging app while Omni Factory works alongside it.

1. Prepare your workspace. Install Omni Factory on a Windows 11 computer. Connect cloud models with your own API keys or set up local models. To use agent suggestions, give the agent a role, a task, and your project materials.

2. Start the conversation and turn on the features you need. Recording and speech recognition turn your conversation into a transcript. Translation helps you communicate across languages. The agent uses your conversation and materials to suggest answers, flag key points, or keep a presentation on track. Turn on only what you need.

3. Review the results after your meeting. Summaries, transcripts, and project history help you revisit agreements and pick up where you left off. For a deeper look, connect ChatGPT, Claude, or your own agents through MCP.

The Omni Factory app is free for personal use with your own API keys. Third-party model usage is billed separately under your provider's terms.

In other words, a free platform doesn't make third-party APIs free. We'll discuss team plans and implementation separately.

An API key lets Omni Factory connect to your chosen model service. Get the key from your provider, then add it in Omni Factory. This is called BYOK: Bring Your Own Key.

The provider sets its own API pricing and limits. Never publish your API key or share it in a message: it gives access to your account with that service.

Click “Request a build” to email our team about the Windows version. Tell us which version of Windows you use and what you'd like to try first.

macOS and Linux versions are in development.

No. Keep using Zoom, Microsoft Teams, Google Meet, or your usual messaging app. Omni Factory works at the system level, so there's no need to move the conversation to a separate room.

Use what you need: recording and transcription, translation, or agent suggestions based on your materials.

Store conversation history locally and choose what to share with the model. For teams, we can set up syncing with the Omni Factory cloud or your corporate cloud.

Local storage and local processing are not the same thing. When you use a cloud model, your chosen provider processes the context you send. We agree on data policies separately for teams.

Omni Factory runs on any Windows 11 desktop or laptop. You don't need a powerful computer to use cloud models. Processing happens in the cloud — you just need an internet connection.

For local models, the hardware you need depends on the models and features you choose. Text, voice, and avatars each place different demands on your computer. Match local models to your hardware. Mix local and cloud processing to fit your needs.

No. Bring your own API keys, run models locally, or use models in your cloud. Choose based on your needs, quality, cost, and acceptable response times.

Let Omni Factory handle model selection and infrastructure setup.

AI can mishear words, miss context, or draw the wrong conclusion. Check important names, numbers, commitments, and decisions against the recording and your source materials.

The agent suggests — you decide. The agent helps you keep track of context. Your expertise and responsibility for the conversation remain yours.

You'll have a summary, transcript, and project history. Organize conversations and materials into projects so you can revisit decisions, agreements, and open questions before the next meeting.

For a deeper look, use ChatGPT, Claude, or your own agents through MCP. Share only the context you're comfortable sending to that service.

No. Start with recording, translation, or agent suggestions. Add voice and avatars when they make sense for your needs.

Keep your voice during translation or conversations with an agent, choose another voice, or add your likeness or a character. Processing runs locally or in the cloud, depending on the model and service you choose. Only use someone else's voice or likeness with their permission.

The team version adds corporate workspaces, shared conversation history, a company knowledge base, and agents tailored to your processes. Connect your systems and materials, manage access, and set your data policies.

The team version isn't just a set of individual installations. We'll agree on the solution and implementation terms based on your organization's needs.

Yes. We deploy and support infrastructure for agents, voice style transfer, and avatars on your servers, in your corporate cloud, or in the Omni Factory cloud.

We choose models and resources for your workload, budget, and response-time requirements. We optimize token and compute use while maintaining the quality you need. Let's discuss what that looks like for your work.