Skip to main content

AI-Starter 2.4.16

· 3 min read
Takaaki Kakei
Takaaki Kakei
Owner

We have released the latest version 2.4.16 of AI-Starter.

Main Update Contents​

  1. Support for Claude 4 series
  2. Started supporting Azure AI Foundry models
  3. Enhanced Gemini-related features
  4. Other minor bug fixes and feature improvements

1. Introduction of Claude 4 Series​

We have introduced the Claude 4 series. The following new AI assistants have been added:

  • Claude Sonnet 4
  • Claude Sonnet 4 Thinking
  • Claude Opus 4

Please note that the following models are no longer supported as Claude Sonnet 4 serves as their superior replacement:

  • Claude 3.7 Sonnet
  • Claude 3.7 Sonnet Extended thinking mode
  • Claude 3.5 Sonnet v2

Assistants using these models will continue to be available, but future updates may cause malfunctions. We recommend migrating to the new models, which offer superior performance and cost-effectiveness.

v2.4.16 Modelsv2.4.16 Models

2. Started Supporting Azure AI Foundry Models​

We have started supporting some models from Azure AI Foundry. Azure AI Foundry is a new platform that centralizes the execution infrastructure for generative AI on Azure and the functions and tools necessary for developing and managing generative AI applications.

https://learn.microsoft.com/en-us/azure/ai-foundry/what-is-azure-ai-foundry

Previously, "Azure OpenAI" was provided as Azure's AI platform, but integration into Azure AI Foundry may progress in the future. Therefore, AI-Starter has started supporting some Azure AI Foundry models first.

Azure AI Foundry offers over 1,900 models. For example, xAI's Grok 3 and Microsoft's Phi-4 are available.

https://azure.microsoft.com/en-us/products/ai-model-catalog#Models

This time, we support xAI's Grok 3 as an experimental AI-Starter assistant. Other models may also be supported without requiring app updates in some cases. If you would like to use them, please contact our support desk.

First, we have added support for the latest version of Gemini 2.5 Flash Preview (May 20, 2025 version). If you have already enabled the Gemini 2.5 Flash Preview assistant, you can use the latest version by changing to the following model:

  • gemini-2.5-flash-preview-05-20

We have also added support for Gemini's image generation feature. You can use it with the following assistants: If you have already enabled the Gemini 2.0 Flash Experimental assistant and would like to use the image generation feature, configuration changes are required, so please contact our support desk. Please note that since these are preview and experimental versions, operation may be unstable.

  • Gemini 2.0 Flash Preview Image Generation
  • Gemini 2.0 Flash Experimental
v2.4.16 Modelsv2.4.16 Models

Furthermore, we have added support for new image models. While the Imagen series previously did not support Japanese, we have confirmed that Japanese input is now possible for Imagen 3 and others following Google's recent updates.

  • Imagen 4
  • Imagen 4 Ultra

Conclusion​

Thank you for reading to the end. AI-Starter is committed to continuous improvement driven by feedback from our users. Please review the contents of this update, and if you have any opinions or requests, we encourage you to share them with us. Your feedback contributes to the further evolution of AI-Starter. Please look forward to future updates.