> ## Documentation Index
> Fetch the complete documentation index at: https://aether.baimoqilin.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Supported AI Models and LLM Providers in Aether

> Aether supports OpenAI, Anthropic, Google, DeepSeek, Groq, Mistral, Cloudflare, and more. Learn how to add providers and switch models.

Aether supports a wide range of LLM providers via the Pi framework that powers it under the hood. You bring your own API keys — nothing is proxied through Aether's servers. Your keys go directly from your device to each provider's API endpoint.

## Supported Providers

| Provider                  | Notes                                                           |
| ------------------------- | --------------------------------------------------------------- |
| **OpenAI**                | GPT-4o, o1, o3, and all other OpenAI chat models                |
| **Anthropic (Claude)**    | Claude 3.5/3.7 Sonnet, Claude Opus, and the full Claude family  |
| **Google (Gemini)**       | Gemini 1.5 Pro/Flash, Gemini 2.0, and other Gemini models       |
| **DeepSeek**              | DeepSeek-V3, DeepSeek-R1                                        |
| **Groq**                  | Ultra-fast inference for Llama, Mixtral, and Gemma models       |
| **Mistral**               | Mistral Large, Mistral Small, Codestral, and others             |
| **Cloudflare Workers AI** | Serverless AI inference at the edge via your Cloudflare account |

<Note>
  The Pi framework supports additional providers beyond those listed here. If your preferred provider is compatible with Pi, it may already be available when you tap **Add Provider** in Settings.
</Note>

## Adding a Provider

<Steps>
  <Step title="Open Settings">
    Tap the **gear icon** in the top-right corner of the chat screen to open Settings.
  </Step>

  <Step title="Go to Providers">
    Tap **Providers** in the Settings menu to see your currently configured providers.
  </Step>

  <Step title="Add a new Provider">
    Tap **Add Provider** and select your desired provider from the list.
  </Step>

  <Step title="Enter your API key">
    Paste or type your API key into the key field. You can obtain keys from each provider's developer console.
  </Step>

  <Step title="Save">
    Tap **Save**. Aether immediately validates the key and makes that provider's models available in the model selector.
  </Step>
</Steps>

## Switching Models

Tap the **model selector** displayed above the composer input bar. The list shows every model available across all of your configured providers. Tap any model to switch to it — the change takes effect on the next message you send.

## Model Configuration

For models that expose reasoning controls — such as OpenAI's o-series (o1, o3) — Aether lets you adjust the **reasoning effort** level. After selecting an o-series model, tap the model name or the configuration icon to reveal the effort slider and choose **low**, **medium**, or **high** based on the trade-off between speed and depth of reasoning you need.

<Note>
  API keys are stored only on your device within Aether's private app storage. They are never sent anywhere other than the respective provider's own API endpoint.
</Note>

<Tip>
  Use **DeepSeek** or **Groq** for fast, cost-effective responses on straightforward tasks. Reach for **Claude** or **GPT-4o** when you need deep reasoning, nuanced writing, or complex multi-step problem solving.
</Tip>