> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ortavox.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ortavox.ai/_mcp/server.

# LLM

> Supported LLM providers and model selection

Ortavox supports multiple leading AI providers out-of-the-box. You select the LLM provider and model as part of the `model` configuration when creating or updating your Agent.

## Supported Providers & Models

### OpenAI

```json
{ "provider": "openai", "model": "gpt-4o-mini" }
```

Recommended starter model. Other popular options: `gpt-4o`, `gpt-4-turbo`.

---

### Anthropic

```json
{ "provider": "anthropic", "model": "claude-3-5-sonnet-20241022" }
```

Excellent at following nuanced instructions and long-context reasoning.

---

### Google

```json
{ "provider": "google", "model": "gemini-1.5-flash" }
```

Strong multilingual performance and fast responses.

---

### Groq

```json
{ "provider": "groq", "model": "llama-3.3-70b-versatile" }
```

Best-in-class speed for latency-sensitive use cases.

---

### DeepSeek

```json
{ "provider": "deep-seek", "model": "deepseek-chat" }
```

Cost-efficient with strong reasoning capabilities.

---

> **Tip**: Keep `temperature` between `0.5` and `0.9` for conversational agents. Lower values make responses more deterministic; higher values create more variability.