Document BYO LLM & BYO Agent

content/en/docs/refguide/mendix-ai-assistance/_index.md

@@ -29,6 +29,8 @@ For more information on how to ensure optimal performance of Maia, see the [Best
 
 Mendix applies fair use limits to maintain a reliable experience for all users. If usage exceeds these limits, your Maia access may be temporarily paused. Account access is reviewed on a weekly schedule. Your account will be automatically reactivated within seven days once the review period concludes.
 
+If you need uninterrupted access to Maia capabilities, you can [configure a custom AI provider](/refguide/maia-make/#custom-provider) to use your own LLM provider.
+
 ## Network Configuration Requirements {#maia-network-requirement}
 
 To ensure the smooth operation of Mendix AI Assistance (Maia), the following network requirements must be met:

content/en/docs/refguide/mendix-ai-assistance/maia-make/_index.md

@@ -102,6 +102,51 @@ The conversational interface includes the following options:
 * **{{% icon name="paperclip" %}} Add** (Image, Story, PDF) - With this option, you can attach images, PDFs, or user stories to help Maia understand your requirements better.
 * **Add file to Maia Chat** (@ icon) - It allows you to add certain logic or pages to Maia as context. You can also access this option by right-clicking the documents (microflows or pages) in the **App Explorer** and it appears in the context menu.
 
+## Configuring a Custom AI Provider {#custom-provider}
+
+By default, Maia uses the Mendix platform's AI service. You can configure Maia to use your own LLM provider instead (also known as bring your own LLM or BYO LLM).
+
+{{% alert color="info" %}}
+Configuring a custom AI provider is available in Studio Pro 11.12 and above as a beta feature.
+{{% /alert %}}
+
+To enable custom LLM provider configuration, follow these steps:
+
+1. Open the Studio Pro settings.
+2. In the **Preferences** dialog box that opens, go to the **New Features** tab.
+3. Select the **Enable configuring a custom AI provider for Maia** checkbox, then click **OK**.
+
+After enabling this feature, a **Settings** ({{% icon name="cog" %}}) icon appears in the upper-right corner of the Maia pane.
+
+### Provider Configuration {#provider-configuration}
+
+Custom AI provider settings are per-project and local to your workstation. Each app can have its own configuration, and settings are not committed to version control.
+
+#### Mendix Platform {#mendix-platform}
+
+The Mendix platform AI service is the default, recommended option and requires no configuration. Some Mendix platform optimizations may not be available if you use a custom LLM provider instead.
+
+#### Amazon Bedrock {#amazon-bedrock}
+
+To configure Amazon Bedrock as your LLM provider, follow these steps:
+
+1. Click **Settings** ({{% icon name="cog" %}}) in the Maia pane.
+2. Select **Amazon Bedrock** from the **LLM Provider** list.
+3. Create an API key in AWS, then enter the **Base URL** and your **API Key**. Studio Pro stores the API key securely across sessions.
+4. Enter the **Model ID**. You can find available model IDs in your AWS Bedrock console. Mendix recommends Claude Sonnet 4.6 for optimal Maia performance.
+5. Click **Save Config**.
+
+#### OpenAI Compatible {#openai-compatible}
+
+To configure an OpenAI-compatible provider, follow these steps:
+
+1. Click **Settings** ({{% icon name="cog" %}}) in the Maia pane.
+2. Select **OpenAI Compatible** from the **LLM Provider** list.
+3. Enter the **Base URL** for your API endpoint.
+4. Enter the **Model ID**.
+5. Optional: Enter your **API Key**. Studio Pro stores this key securely across sessions.
+6. Click **Save Config**.
+
 ## Save a Maia Session {#save-maia-session}
 
 A Maia session contains your conversation history and context, which can help analyze Maia's responses and behavior for troubleshooting or support purposes. To save a Maia session, go to **Help** > **Support Tools** > **Capture Maia Agent Session**.

content/en/docs/refguide/mendix-ai-assistance/maia-make/studio-pro-mcp-server.md

@@ -15,7 +15,7 @@ This feature was released in Studio Pro 11.10.
 To use Studio Pro MCP Server, an internet connection and signing in to Studio Pro are required.
 {{% /alert %}}
 
-Studio Pro MCP Server enables you to leverage Maia's capabilities directly from external clients, including AI coding assistants, agents, and other MCP-based LLM tools.
+Studio Pro MCP Server enables bring your own agent (BYO Agent) capabilities, allowing you to leverage Maia's capabilities directly from external clients, including AI coding assistants, agents, and other MCP-based LLM tools.
 
 It has the following key highlights: