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

# Getting Started with Tyk AI Studio

> In this guide, we will walk through the initial configuration steps after deploying Tyk AI Studio, including logging in, connecting an LLM provider, and starting your first chat.

## Availability

| Edition | Deployment Type |
| :- | :- |
| [Community](https://tyk.io/docs/ai-management/ai-studio/overview.md#community-edition) & [Enterprise](https://tyk.io/docs/ai-management/ai-studio/overview.md#enterprise-edition) | Self-Managed |

> [!NOTE]
>
> This guide assumes you are logged in as a **Studio Administrator** with full permissions to configure the system.

## 1. Login to the AI Studio

After completing the [installation process](https://tyk.io/docs/ai-management/ai-studio/quickstart.md) and registering your first user:

1. **Access the UI:** Open your web browser and navigate to your configured `SITE_URL`
2. **Admin Login:** Log in using the administrator account you created during registration:

   ![Login Screen](https://mintcdn.com/tyk/KUyxLx5tNlKCB02w/img/ai-management/login-screen.png?fit=max\&auto=format\&n=KUyxLx5tNlKCB02w\&q=85\&s=5f6bc0de58af5bed8a323c0bbfda19a1)

   > [!NOTE]
   >
   > **Reminder**: If you haven't completed the initial registration yet, go back to your [installation guide](https://tyk.io/docs/ai-management/ai-studio/quickstart.md#first-user-registration) and follow the "First User Registration" section.

## 2. Configure Your First LLM

One of the most common initial steps is connecting Tyk AI Studio to an LLM provider.

1. **Navigate to LLM Management:** In the admin UI sidebar, select **LLM management > LLM providers**.
2. **Add LLM Configuration:** Click the button to add a new LLM Configuration.
3. **Enter Details:**
   * **Configuration Name:** Give it a recognizable name (e.g., `OpenAI-GPT-4o`).
   * **Description:** Optionally, add a description for this configuration.
   * **Select Vendor:** Choose the LLM provider you want to connect (e.g., OpenAI, Anthropic, Azure OpenAI).
   * **Select Default Model:** Specify the exact model identifier(s) provided by the vendor (e.g., `gpt-4o`, `gpt-4-turbo`).
   * Add API Key in the **Access Details** section:
     > [!NOTE]
     >
     > Do *not* paste your API key directly here. Instead, use [Secrets Management](https://tyk.io/docs/ai-management/ai-studio/secrets.md).
     * If you haven't already, go to **Settings > Secrets** in the admin UI and create a new secret:
       * **Variable Name:** `OPENAI_API_KEY` (or similar)
       * **Secret Value:** Paste your actual OpenAI API key here.
       * Save the secret.
     * Return to the LLM Configuration screen.
     * In the API Key field, enter the secret reference: `$SECRET/OPENAI_API_KEY` (using the exact Variable Name you created).
   * **Other Parameters:** Configure any other provider-specific settings (e.g., Base URL for Azure/custom endpoints, default temperature, etc.).
4. **Save:** Save the LLM configuration.

   ![LLM Config UI](https://mintcdn.com/tyk/Fkmq9JBq93TelQVC/img/ai-management/initial-configuration-llm-config-ui.png?fit=max\&auto=format\&n=Fkmq9JBq93TelQVC\&q=85\&s=e35100e6cb8d636550ea51494e7b5ebe)

In the Community Edition, AI Studio adds a new LLM vendor to the Default catalog when you save it. The vendor is then available to [users](https://tyk.io/docs/ai-management/ai-studio/user-management.md) who have access to the Default catalog. In the Enterprise Edition, AI Studio does not do this. Add the vendor to a [catalog](https://tyk.io/docs/ai-management/ai-studio/catalogs.md) that the correct Teams have.

For more details, see the [LLM Management](https://tyk.io/docs/ai-management/ai-studio/llm-management.md) documentation.

## 3. Create Chat Experience

1. Navigate to **Chat > Chats** in the admin UI sidebar.
2. Click to create a new Chat Experience.
   * **Name:** Give your chat a descriptive name (e.g., `OpenAPI GPT4o`).
   * **LLM Call settings:** Select the [LLM Call settings](https://tyk.io/docs/ai-management/ai-studio/call-settings.md) you want to use for this chat (e.g., `gpt-4o`). If the specific model settings are not available, you can create one in the [LLM model settings](https://tyk.io/docs/ai-management/ai-studio/call-settings.md#how-to-create-llm-call-settings) page. We have added defaults for the popular ones.
   * **Select LLM vendor:** Select the LLM configuration you created in the previous step.
   * **Select Group**: Assign this chat to a specific Group or select "Default" to make it available to all users.
3. Save the new Chat Experience.

   ![Chat Config](https://mintcdn.com/tyk/hBFyexPCuAxneHvv/img/ai-management/chat-experience-config.png?fit=max\&auto=format\&n=hBFyexPCuAxneHvv\&q=85\&s=335a1389c3effaba0ffe97936a7e0231)

## 4. Use the Chat Interface

1. Select the **Chat** tab at the top of the UI.
2. **Select Chat Experience:** Choose the `OpenAPI GPT4o` experience you just created from the list of available chats.
3. **Interact:** Type a question or prompt in the chat box.
4. **Receive Responses:** The LLM will process your request and stream the response back to you in the chat window.

   ![Chat UI](https://mintcdn.com/tyk/KUyxLx5tNlKCB02w/img/ai-management/chat-interface-ui.png?fit=max\&auto=format\&n=KUyxLx5tNlKCB02w\&q=85\&s=6bd25d47a94bc0b7f8c890cfa4e79ba3)

You have now successfully configured an LLM and used it to answer a question through the Tyk AI Studio Chat Interface.

## Next Steps

With the initial configuration complete, you can now:

* Explore [User Management](https://tyk.io/docs/ai-management/ai-studio/user-management.md) to create users and groups.
* Set up [Tools](https://tyk.io/docs/ai-management/ai-studio/tools.md) for external API integration.
* Configure [Data Sources](https://tyk.io/docs/ai-management/ai-studio/datasources-rag.md) for RAG.
* Define [Filters](https://tyk.io/docs/ai-management/ai-studio/filters.md) for custom request/response logic.