Version 1.3

Version 1.3 supports all features of 1.2.

Added support for two newer language models, GPT-4o and GPT-4o Mini.

• In order to upgrade, GPT-4o APIM endpoints should be added under the existing root API OpenAI4: /openai4/chat4o and /openai4/chat4omini

• Parallel function calling, image generation and recognition are seamlessly supported for new models.

• Updated documentation with these new models.

Version 1.2

Release 1.2 supports all features of 1.1.

New features and improvements:

1. Data encryption for all storage types.

• SharePoint list
• Database
• Local Storage
• Seamless support of the Chat Sharing option for all storage types
• Max chat name length is limited to 150 chars for encrypted storages and 255 for plain ones.

2. GPT-4 Vision APIM endpoint (/openai4/vision).

• Prerequisites:
  • To enable the deployment of Vision-preview in Azure Open AI, you must configure the Azure AI Computer Vision resource in the same Azure region as your GPT-4 with Vision resource.
    • More information: https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/gpt-with-vision
    • By default, this resource uses Price Tier: Free (F0), which supports 20 Calls per minute, 5K Calls per month.
    • You can choose the more capable Price Tier: S1, which supports 10 Calls per second and costs ~1$ per 1000 calls.
  • The recommended zone for the deployment is Swedish Central (as of Dec 2023). It supports both types of imaging models Vision and Dalle3 for image generations.
• Optional checkbox to enable Vision model under the enabled integrations.

3. Speech synthesis to read out AI-generated texts by default

   • The standard Web Speech API requires selecting the preferred language; using default page language is not always optimal. This default option is used when Native OpenAI text-to-speech model is not available.
   • Azure OpenAI does not yet have support for text-to-speech models (as of Dec 2023).

4. /tts APIM endpoint (/openainative/tts) and Native OpenAI text-to-speech model for AI-generated texts.

   • In case of using Native OpenAI endpoints - direct URL or APIM-based operation - the native text-to-speech model automatically handles text that contains mixed languages.

Bug fixes
a. Revised the PDF summarization logic to suppress content errors related to rapidly exhausted max content length limits in old models like GPT 3.5 Turbo 16k.

• The new logic does not send previously processed PDF content of the same chat when the new PDF document has been loaded for processing.

b. Optimized image compacting logic for images uploaded to be analyzed with GPT-4 Vision model.

• Uploaded images are saved to Chat Images if the library exists.
• Otherwise uploaded images are compressed to JPEG format (as in Version 1,1).

Version 1.1

Release 1.1 supports all key features of 1.0.

New features and improvements:

1. The option to search on the internet with Azure Bing Search service.
2. The option to search on the internet with Google Custom Search service.
3. The option to generate images with the Dalle 3 model from prompt text like "Create an image: an illustration of a business meeting."
4. Improved performance for event streaming on a large number of own chats in the navigation.
5. Added the Stop streaming button for the upper and lower positions of the prompt text area.
6. Minor changes to the Code Highlighter.
7. Added support for the new Azure OpenAI model GPT-4-1106-preview (GPT-4 Turbo).

• It supports 128k input tokens.
• The preview model is available for the selection in the web part settings. It is disabled by default.

8. The option to display the dropdown box with examples for prompt text.
9. The option to use voice input to prompt text. It supports 16 popular languages.

Fixed bugs of the previous version

1. Slow rendering when the Code Highlighting option enabled. The rendering logic was optimized,
2. Image analysis with GPT-4 Vision: saving Chat history could fail if the user selected a large image generated by Dalle3 for the analysis. Image saving logic was optimized.
3. Requests failed when Native OpenAI was used with disabled Event Streaming option. The logical mistake fixed.
4. If SPFx permissions have not been approved, the request logic stuck on attempts to execute People Search and select people from People Picker for sharing. The logic has been improved.

Version 1.0

Key features

• Supports Azure OpenAI, and optionally, Native OpenAI
• Chat history with 3 storage options:
  • SharePoint Custom List (default)
  • Azure SQL Database
  • Local Browser Storage (for quick demo)
• Global and Private Chat Sharing for selected Azure AD accounts
• Default support for GPT 4 32k and GPT 3.5 16k Text Models for Azure OpenAI
• Support for any available text models for Native OpenAI, including (as of November 2023):
  • gpt-4-vision-preview: Image Analysis: Upload images to describe their Content
  • gpt-4-1106-preview
• Integration with Azure API Management Service
• Optional integrations with company data:
  • SharePoint Search
  • Company Users
  • Local Date and Time
  • Analysis of an uploaded PDF and summarization of its content
  • Analysis of uploaded images and description of their content
• Code Highlighting with Configurable Styles
• Consecutive Event Streaming
• Full-Screen Mode
• Unlimited Length of Chat History (Configurable)
• Upper and Lower side positioning for the Prompt Text Area
• Configurable Formats for Dates: Default is Finnish, which can be changed to "en-US" in Web Part Settings.

The recently added new features of version 1.1 are described here.