Customer-facing AI
Add AI inside your own product
This is separate from the AI developer in Studio. When a site, bot or app needs AI for its users, Studio connects Gemma through Google AI Studio.
- Windows
- Linux AppImage
- Linux .deb
- macOS coming soon
Windows 10 or later or Ubuntu 22.04 or later (64-bit). macOS coming soon. About 1 GB of free disk space.
Windows may show a SmartScreen warning: click “More info”, then “Run anyway”.
Updated: .
What project AI can do
Understand multiple media types
Text, images, audio and video in one workflow.
Answer from documents
RAG and GraphRAG stay under the hood when answers must be grounded in project material.
Work with data
The project can read permitted SQL data and store the required result.
Keys and limits
In managed mode the Google key stays on Dialektai servers and never enters public client code. Studio applies limits, timeouts, retries and a clear fallback if the model is unavailable.
Questions
DeepSeek or Gemma?
DeepSeek powers development work in Studio. Gemma is used when the user project itself needs AI.
Should a visitor receive the API key?
No. Secrets must never be embedded in client code and stay on the server side.
Describe what AI should do for a customer.
Studio will propose a safe architecture inside the project.