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

# AI Providers

> Learn how to integrate and manage API Keys for AI providers like OpenAI, Gemini, Claude, DeepSeek, OpenRouter, and OpenAI Compatible in ChatbotX.

To enable your AI Agent to think and respond to customers intelligently and naturally, you need to connect your ChatbotX workspace to one or more artificial intelligence providers (**AI Providers**). ChatbotX supports direct API integrations with the world's leading Large Language Model (LLM) platforms.

Integrating AI Providers helps you:

* **Gain Flexibility:** Choose the LLM that best fits your budget and the specific conversational needs of your business.
* **Automate Interactions:** Enable your bot to study your Knowledge base and answer customer queries automatically 24/7.
* **Optimize Costs:** Pay for AI usage directly to the provider based on your actual token consumption without any markup fees.

## AI Provider Configuration Page

You can manage your AI connections by navigating to your admin Portal, selecting **Settings** from the left navigation menu, and clicking the **Integrations** tab on the top toolbar.

<Frame>
  <img src="https://mintcdn.com/chatbotx/IGA3VdhbQlsJyII7/images/chatbotx_settings_integrations_ai_providers.png?fit=max&auto=format&n=IGA3VdhbQlsJyII7&q=85&s=c1f25b3ddef25c3e5072cf283aae9472" alt="Chatbotx Settings Integrations AI Providers" width="2808" height="1260" data-path="images/chatbotx_settings_integrations_ai_providers.png" />
</Frame>

## Integrating Your LLM Accounts

ChatbotX supports the major AI providers listed below. Follow the quick guide for each platform to retrieve and link your API Keys:

### 1. OpenAI

The most popular AI platform, providing highly intelligent and reliable responses.

* **How to integrate:**
  1. Log in to your OpenAI Developer Dashboard.
  2. Create a new API Key and copy it.
  3. Expand the **OpenAI** section under the Integrations tab in ChatbotX, paste your API Key, and click connect.
  4. Configure the **Automated Responses** option:
     * **When Enabled:** Every time a user sends a message to your chatbot, the chatbot will automatically use the AI model to analyze your knowledge base and respond to the user's message (Global Auto Reply).
     * **When Disabled:** Global automated responses are disabled. You can still use OpenAI capabilities by manually configuring OpenAI actions within your chatbot flows (**Flow Builder**).
* **API Key URL:**
  ```text theme={null}
  https://platform.openai.com/api-keys
  ```

### 2. Google Gemini

Features exceptionally fast processing speeds and a massive context window for handling large documents.

* **How to integrate:**
  1. Access Google AI Studio.
  2. Click **Get API key** and generate a new key.
  3. Paste this key into the **Google Gemini** section in the ChatbotX dashboard.
* **API Key URL:**
  ```text theme={null}
  https://aistudio.google.com/app/apikey
  ```

### 3. Claude AI

AI models by Anthropic, renowned for superior logical reasoning and highly professional, natural conversations.

* **How to integrate:**
  1. Access and log in to the Anthropic Console.
  2. Navigate to the **API Keys** section and generate a new key.
  3. Copy and paste the key into the **Claude AI** section on ChatbotX to activate.
* **API Key URL:**
  ```text theme={null}
  https://console.anthropic.com/settings/keys
  ```

### 4. DeepSeek

Next-generation AI models that deliver high performance and exceptional language understanding at highly optimized API costs.

* **How to integrate:**
  1. Log in to the DeepSeek API Dashboard.
  2. Select **API Keys** from the left menu and generate a new key.
  3. Copy the key and paste it into the **DeepSeek** section in ChatbotX.
* **API Key URL:**
  ```text theme={null}
  https://platform.deepseek.com/api_keys
  ```

### 5. OpenRouter

A unified router gateway that gives you access to hundreds of open-source and commercial AI models through a single API Key.

* **How to integrate:**
  1. Log in to your OpenRouter account.
  2. Create a new API Key in your key management settings.
  3. Paste this key into the **OpenRouter** section in ChatbotX, then select your desired LLM from the connected dropdown list.
* **API Key URL:**
  ```text theme={null}
  https://openrouter.ai/keys
  ```

### 6. OpenAI Compatible

**OpenAI Compatible** allows you to connect any AI provider that uses the OpenAI-compatible API format. Instead of being limited to the built-in providers, you can configure custom connections to other AI services, self-hosted models, or any platform that exposes an OpenAI Chat Completions-compatible endpoint.

<Steps>
  <Step title="Open OpenAI Compatible section">
    Go to **Settings** > **Integrations** > Expand the **OpenAI Compatible** section and click **+ Add AI Provider**.

    <Frame>
      <img src="https://mintcdn.com/chatbotx/D8JmjgGdF4aB5Ksr/images/openai_compatible_integration_settings.png?fit=max&auto=format&n=D8JmjgGdF4aB5Ksr&q=85&s=816aee1a15b73fce9edf9b7e0532fd87" alt="Openai Compatible Integration Settings" width="2892" height="1160" data-path="images/openai_compatible_integration_settings.png" />
    </Frame>
  </Step>

  <Step title="Select a Provider or create Custom">
    In the **Add AI Provider** modal, select a provider from the preset list (NVIDIA NIM, LM Studio, Heroku, Clarifai, NEAR AI Cloud...) or choose **Custom** to configure your own.

    <Frame>
      <img src="https://mintcdn.com/chatbotx/D8JmjgGdF4aB5Ksr/images/openai_compatible_add_provider.png?fit=max&auto=format&n=D8JmjgGdF4aB5Ksr&q=85&s=50da14d8a82dc750bfe7c1d40ce2d05c" alt="Openai Compatible Add Provider" width="2921" height="1483" data-path="images/openai_compatible_add_provider.png" />
    </Frame>
  </Step>

  <Step title="Fill in connection details">
    Configure the following fields:

    * **Name**: A display name to distinguish between providers (e.g., "LM Studio Local", "Ollama Server").
    * **Base URL**: The API endpoint address of the provider (see the reference table below).
    * **Default model**: The default model name to use (e.g., `gpt-4o-mini`, `llama-3.1-8b`).
    * **API Key**: The authentication key from the provider (some local services like LM Studio or Ollama may not require one).
    * **AI Auto Reply**: Enable to use this provider for the global auto-reply feature.
    * **Enabled**: Toggle on to activate this provider.

    <Frame>
      <img src="https://mintcdn.com/chatbotx/D8JmjgGdF4aB5Ksr/images/openai_compatible_custom_config.png?fit=max&auto=format&n=D8JmjgGdF4aB5Ksr&q=85&s=6367f3854fe688f09e333b3328cbed1a" alt="Openai Compatible Custom Config" width="2919" height="1520" data-path="images/openai_compatible_custom_config.png" />
    </Frame>
  </Step>
</Steps>

Click **Confirm** to save and activate the connection.

#### Common Base URL Reference

| Provider / Service | Base URL                                                              | Notes                                                           |
| ------------------ | --------------------------------------------------------------------- | --------------------------------------------------------------- |
| Together AI        | `https://api.together.xyz/v1`                                         | Wide range of open-source models                                |
| LM Studio (Local)  | `http://localhost:1234/v1`                                            | Run AI models on your local machine                             |
| Ollama (Local)     | `http://localhost:11434/v1`                                           | Run open-source models locally                                  |
| NVIDIA NIM         | `https://integrate.api.nvidia.com/v1`                                 | GPU-optimized models from NVIDIA                                |
| Groq               | `https://api.groq.com/openai/v1`                                      | Ultra-fast inference speed                                      |
| Fireworks AI       | `https://api.fireworks.ai/inference/v1`                               | Fast open-source model deployment                               |
| Mistral AI         | `https://api.mistral.ai/v1`                                           | Mistral, Mixtral models                                         |
| Perplexity         | `https://api.perplexity.ai`                                           | AI with web search capabilities                                 |
| Azure OpenAI       | `https://{resource}.openai.azure.com/openai/deployments/{deployment}` | Replace `{resource}` and `{deployment}` with your Azure details |

<Note>
  You can add multiple OpenAI Compatible providers at the same time. Each provider will appear as a separate option when configuring the AI model for an AI Agent.
</Note>

## Assigning an AI Provider to an AI Agent

Once you have successfully linked your API Keys in Settings, proceed to configure the specific AI model and parameters for each AI Agent:

1. Go to the **AI Agents** menu on the left sidebar.
2. Select the AI Agent you want to configure or click **+ Create AI Agent** to build a new one.
3. In the agent configuration form, set up the parameters on the right panel to allocate the LLM model for your agent:

<Frame>
  <img src="https://mintcdn.com/chatbotx/VgyR1lnh4HUf8xti/images/chatbotx_ai_agents_configure_llm_models.png?fit=max&auto=format&n=VgyR1lnh4HUf8xti&q=85&s=a1cf11d0c7a549b47316baf452eb79ac" alt="Chatbotx Ai Agents Configure Llm Models" width="2918" height="1815" data-path="images/chatbotx_ai_agents_configure_llm_models.png" />
</Frame>

* **Model selection (Gemini Model / OpenAI Model / Claude Model / DeepSeek Model / OpenRouter Model / OpenAI Compatible):** Choose the specific LLM version you want to run for each provider integrated under Settings. Any OpenAI Compatible providers you have added will also appear in this selection list.
* **Temperature:** A slider that controls the randomness and creativity of the AI's responses (valued from `0` to `1`).
  * *Low Values (e.g., 0.2 - 0.4):* The bot responds with factual, precise answers strictly following your Knowledge base (ideal for Customer Support or Sales).
  * *High Values (e.g., 0.7 - 0.9):* The bot responds with more creative, diverse, and expressive language (ideal for copywriting or content generation).
* **Max Tokens:** Limits the maximum size of the bot's response (e.g., `2048`) to prevent verbose outputs and help optimize your API usage costs.

Click **Confirm** (or **Save**) to apply the configuration to the AI Agent.