> For the complete documentation index, see [llms.txt](https://docs.siminsights.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.siminsights.com/hyperskill-studio-authoring/ai-features/large-language-models-and-data-privacy.md).

# Large Language Models & Data Privacy

AI Chat and Agents are powered by large language models (LLMs). This page explains which models you can choose and where to find each provider's data policy.

## Choose Your AI Model

When you use HyperSkill, you can choose between several LLMs to power AI Chat and Agents: GPT-4, GPT-4o, GPT-4o mini, GPT-5 and Gemini. Both AI Chat and Agents let you choose which model to use, and you are free to mix models across them. The list you see can differ slightly between AI Chat and Agents.

Each model has its own strengths, so the best choice depends on your project:

* **GPT-4 and Gemini:** these tend to respond faster.
* **GPT-5:** it has better reasoning, but takes longer to run.

## Data privacy

### GPT models (Azure)

All GPT models run on Azure. See Microsoft's [Azure OpenAI data privacy policy](https://learn.microsoft.com/en-us/legal/cognitive-services/openai/data-privacy).

### Google Gemini (free)

See [Google Gemini's data policies](https://ai.google.dev/terms#content-license).

### Enterprise

For enterprise and privacy needs, contact <info@siminsights.com> to learn about HyperSkill's data security solutions.