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

# Integrations

> Connect exchanges, AI model providers, and messaging platforms to Gigabrain

Gigabrain integrates with third-party platforms for trade execution and real-time notifications. All integrations are managed from your [Profile → Integrations](https://gigabrain.gg/profile?tab=integrations) page.

<Warning>
  Gigabrain is non-custodial. We never have access to your funds. Exchange API keys only allow Gigabrain to place orders you configure, it cannot withdraw funds. Never share your wallet seed phrase or recovery phrase through the platform.
</Warning>

## Hyperliquid

Hyperliquid is the primary execution venue for Gigabrain Agents. Connecting it allows your Agents to place, monitor, and close trades on your behalf based on the strategies you define.

### Setup

<Steps>
  <Step title="Visit integrations page">
    1. Go to [Profile → Integrations](https://gigabrain.gg/profile?tab=integrations).
    2. Select **Hyperliquid**.
    3. Connect your wallet and sign transactions to authorize generate apis wallet to trade on your behalf.
  </Step>

  <Step title="Verify connection">
    Once connected, you should see a **Connected** status on your Integrations page. Your Agents can now execute trades on Hyperliquid.
  </Step>
</Steps>

### Key considerations

* **Execution risk**: Orders are submitted to Hyperliquid under its rules. Slippage, partial fills, and liquidation mechanics are controlled by Hyperliquid, not Gigabrain. See the [Risk Disclosure](/support/risk-disclosure) for details.
* **Use Standard Hyperliquid account**: For new account Hyperliquid will create a unified account for you. At momentum we don't support unified accounts yet. You can diable the unified account in hyperliquid settings, by clicking on "Disable Unified Account Mode".

## Telegram Notifications

Connect Telegram to receive real-time alerts from your Agents, including trade executions, performance summaries, and market signals.

### Setup

<Steps>
  <Step title="Visit integrations page">
    1. Go to [Profile → Integrations](https://gigabrain.gg/profile?tab=integrations).
    2. Select **Telegram**.
    3. Copy the displayed code and message it to [telegram bot](https://t.me/AskGigabrainBot).
  </Step>

  <Step title="Verify connection">
    1. You should see a **Connected** status on your Integrations page.
  </Step>
</Steps>

## AI Models (Bring Your Own Key)

Connect your own LLM provider API key to use any model from OpenAI, Anthropic, xAI, or OpenRouter with Gigabrain Chat and Agents. When using your own key, only the base platform fee is charged — no token markup.

### Supported providers

| Provider       | Models                                      | How to get a key                                                     |
| -------------- | ------------------------------------------- | -------------------------------------------------------------------- |
| **OpenRouter** | 400+ models from all providers with one key | [openrouter.ai/keys](https://openrouter.ai/keys)                     |
| **OpenAI**     | GPT and reasoning models                    | [platform.openai.com/api-keys](https://platform.openai.com/api-keys) |
| **Anthropic**  | Claude family of models                     | [console.anthropic.com](https://console.anthropic.com/)              |
| **xAI**        | Grok family of models                       | [console.x.ai](https://console.x.ai/)                                |
| **Venice**     | Privacy-focused AI inference                | [venice.ai](https://venice.ai/)                                      |

### Setup

<Steps>
  <Step title="Add your API key">
    1. Go to [Profile → Integrations](https://gigabrain.gg/profile?tab=integrations).
    2. Select **AI Models**.
    3. Choose a provider and paste your API key.
  </Step>

  <Step title="Select a model">
    In Chat, click the model selector (bottom of the input box) to browse available models from your connected providers.
  </Step>
</Steps>

### Using custom models via API

Pass `model` and `model_provider` in your `/v1/chat` request to use a specific model:

```json theme={null}
{
  "message": "Analyze ETH technicals on the 4H chart",
  "model": "claude-sonnet-4-20250514",
  "model_provider": "anthropic"
}
```

For OpenRouter models, use the full OpenRouter model ID and set provider to `openrouter`:

```json theme={null}
{
  "message": "What's the macro risk regime?",
  "model": "anthropic/claude-sonnet-4",
  "model_provider": "openrouter"
}
```

### Key considerations

* **API keys are encrypted at rest** using AES-256-GCM. They are never stored in plain text.
* **One key per provider.** Adding a new key for the same provider replaces the existing one.
* **Fallback to default.** If no custom model is selected, Gigabrain uses its default model.
* **OpenRouter is the easiest way to start.** One key gives you access to models from OpenAI, Anthropic, Google, Meta, and more.

## Managing Integrations

* View and manage all connected integrations from your [Profile → Integrations](https://gigabrain.gg/profile?tab=integrations) page.
* You can disconnect any integration at any time.
* Disconnecting an exchange integration will stop any active Agents from executing trades.

Need help? Contact on [Telegram](https://t.me/gigabraingg).