Azure OpenAI
Azure OpenAI Service is a managed service that allows developers to integrate OpenAI models into their applications easily:
Access pre-trained models such as natural language processing, computer vision, and more.
Fine-tune the included pre-trained models with your organizational IP for a specialized model.
For more information on the Azure OpenAI service, please see the official site and blog posts.
The Azure OpenAI GenAI Agent allows you to query your OpenAI model securely hosted in Azure, and include the result in the Data Stream.
The Azure OpenAI Embedding GenAI Agent allows you to convert input text into an embedding vector for use in RAG-based retrieval and similarity searches.
Pre-requisites
The following is required to use the Azure OpenAI Agent:
Azure OpenAI instance credentials are required to use this Agent.
Azure OpenAI v1.05 and earlier require an XMPro Stream Host (SH) below v4.5.0. Use v1.06 for compatibility with all SH versions.
The following is required to use the Azure OpenAI Embedding Agent:
A deployment of Azure OpenAI with access to embedding models.
Current Version
Request the Azure OpenAI GenAI Agent v2.01 (Last Update: 28 May 2026)
Request the Azure OpenAI Embedding GenAI Agent v1.00 (Last Update: 18 Nov 2025)
Azure OpenAI Release Notes
2.01
28 May 2026
Repackaged to fix the missing configurable API timeout.
2.00
19 May 2026
Support Event-Level Streaming (when using XMPro Platform v4.6+) for reduced memory usage and earlier downstream processing.
Support a configurable API timeout so that requests do not hang indefinitely under slow or unreliable network conditions.
Support token usage as output attributes for transparency and reporting. Note: Packaging issues. Use v2.01+ instead.
1.11
06 May 2026
Repackaged to add missing DLL file and drop-down translations.
1.06
10 Nov 2025
Repackaged for compatibility with Stream Host v4.5.0+.
1.05
23 Dec 2024
Repackaged to translate the Agent's properties and change the category to Generative AI.
1.04
09 May 2024
Support a dynamic message type.
Support max tokens to remove messages when the defined max is reached.
Removed the [ASSISTANT] prefix so that the output payload contains only the OpenAI response.
1.02
05 Mar 2024
Repackaged the Agent:
Changed the Category from Action Agent to AI & ML
Renamed the Message's Friendly Name to Description
Reworded the Description
1.01
22 Feb 2024
Added a friendly name to the Message List.
1.00
20 Feb 2024
Initial Release.
Azure OpenAI Embedding Release Notes
1.00
18 Nov 2025
Initial Release.
Last updated
Was this helpful?