# AI Integrations

Connect AI providers to power the AI Assistant and module generation.

## Overview

ARI supports multiple AI providers for the built-in AI Assistant and AI-powered module generation. All API keys are stored locally on your server and never shared with ARI or any third party.

Configure your AI providers in **Settings \> Integrations**. API keys you enter through the UI are private to your account — they are not shared with other user accounts on your ARI instance.

When you add API keys in **Settings \> Integrations**, they are stored in your database and encrypted using AES-256-GCM.

Alternatively, you can store your API keys in your `.env.local` file or your host's [environment variables manager](https://vercel.com/docs/environment-variables/managing-environment-variables). This reduces risk because even if your database is ever accessed or compromised, your API keys are not included. It also limits exposure within your application, since the keys are not retrievable through the app itself.

All environment variables added to `.env.local` are shared with all user accounts on ARI — every user of your instance uses the same key. Keys entered through **Settings \> Integrations**, by contrast, are private to the account that enters them.

Choose the approach that aligns best with your security preferences and environment setup.

![ARI Settings page on the AI Providers tab showing API key fields and model overrides for OpenRouter, Claude, OpenAI, and Google Gemini](https://ari.software/ARI-AI-Providers.png)

## Supported Providers

### OpenAI

The default provider for the AI Assistant module and module generation.

1. Create an account at [platform.openai.com](https://platform.openai.com)
2. Navigate to API Keys and generate a new key
3. Add it in Settings \> Integrations or set `OPENAI_API_KEY` in your environment

### Anthropic (Claude)

Use Claude models for the AI Assistant.

1. Create an account at [console.anthropic.com](https://console.anthropic.com)
2. Generate an API key
3. Add it in Settings \> Integrations or set `ANTHROPIC_API_KEY` in your environment

### OpenRouter

Access a wide range of AI models through a single API. OpenRouter provides access to models from multiple providers.

1. Create an account at [openrouter.ai](https://openrouter.ai)
2. Generate an API key
3. Add it in Settings \> Integrations or set `OPENROUTER_API_KEY` in your environment

### Google Gemini

Use Google's Gemini models for the AI Assistant.

1. Get an API key from [Google AI Studio](https://aistudio.google.com)
2. Add it in Settings \> Integrations or set `GOOGLE_GEMINI_API_KEY` in your environment

### xAI (Grok)

Use xAI's Grok models with real-time knowledge.

1. Create an account at [console.x.ai](https://console.x.ai)
2. Generate an API key
3. Add it in Settings \> Integrations or set `XAI_API_KEY` in your environment

### Mistral

Use Mistral's open-weight and frontier models.

1. Create an account at [console.mistral.ai](https://console.mistral.ai)
2. Generate an API key
3. Add it in Settings \> Integrations or set `MISTRAL_API_KEY` in your environment

### DeepSeek

Strong reasoning models at low cost.

1. Create an account at [platform.deepseek.com](https://platform.deepseek.com)
2. Generate an API key
3. Add it in Settings \> Integrations or set `DEEPSEEK_API_KEY` in your environment

### Groq

Ultra-fast inference for open-source models.

1. Create an account at [console.groq.com](https://console.groq.com)
2. Generate an API key
3. Add it in Settings \> Integrations or set `GROQ_API_KEY` in your environment

### Perplexity

Search-augmented LLMs with online sources.

1. Create an account at [perplexity.ai](https://www.perplexity.ai)
2. Generate an API key in your account settings
3. Add it in Settings \> Integrations or set `PERPLEXITY_API_KEY` in your environment

### ElevenLabs

Realistic AI voice synthesis and text-to-speech.

1. Create an account at [elevenlabs.io](https://elevenlabs.io)
2. Generate an API key
3. Add it in Settings \> Integrations or set `ELEVENLABS_API_KEY` in your environment

### Ollama

Run open-source models locally or on a self-hosted endpoint. No API key required.

1. Install Ollama from [ollama.com](https://ollama.com) and pull a model
2. Add your endpoint URL in Settings \> Integrations or set `OLLAMA_BASE_URL` in your environment (default: `http://localhost:11434`)

## Security Best Practices

- Never share your API keys or commit them to version control
- Use different keys for development and production
- Rotate keys periodically
- Set usage limits and billing alerts on each provider's dashboard
- Review API usage in each provider's dashboard regularly

## Your Personal Workspace. Your Rules.

Source Code Available. Self-Hosted. Full Data Control. Fully Extendable. AI Native. Dozens of Built-In Modules And Build Your Own In Minutes. No Coding Required.

[Get Started](https://ari.software/docs/installation)

---

Source: https://ari.software/docs/api-integrations

Markdown rendering of this page, served for `Accept: text/markdown`. Full site index for agents: https://ari.software/llms.txt