Please explain the reasoning behind changing the Gemini Image Tool authentication to use the Gemini API.

[tomy-kyu]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

Content

Please explain why the Gemini Image Tool authentication was changed from using Vertex AI to using the Gemini API.
https://marketplace.dify.ai/plugins/langgenius/gemini_image

I believe that in v0.0.1, it could be used by registering a Vertex AI key, but after the version upgrade, it started requiring a Gemini API key.

We intended to follow Google's recommendation and use the Vertex AI API for the Nano banana model. However, starting from v0.1.0, a separate Gemini API key is required. We were unclear about why this change occurred and whether the Vertex AI API remains usable. Therefore, we are inquiring about the intent behind this change.

In any case, I felt this contradicted the content described below, so I asked the question.

---

The Gemini Image Tool is used to access Gemini's image generation and editing models on Vertex AI. Currently, this plugin supports the following models.

[dosubot[bot]]

The Gemini Image Tool switched from Vertex AI authentication to requiring a Gemini API key starting with version 0.1.0. This change is reflected in both the code and documentation: the tool now makes direct HTTP requests to the Gemini API endpoint and uses the API key in the x-goog-api-key header for authentication, rather than relying on Vertex AI service account credentials or client libraries. All Vertex AI-related logic and configuration were removed from the codebase, and the plugin now only supports Gemini API keys for authentication. There is no indication that Vertex AI authentication remains supported after this change [source][source].

While the pull requests and documentation do not provide an explicit rationale, the change likely reflects a strategic or technical decision to consolidate on the Gemini API for image generation and editing models. This may be due to updated Google endpoints, authentication practices, or differences in access and policy between Vertex AI and Gemini API. The Vertex AI plugin was also removed from the repository, further indicating that Vertex AI support has been discontinued in the plugin ecosystem [source].

If you see any references to Vertex AI in the documentation, those are likely outdated. The current tool implementation, configuration, and credential validation are all designed for Gemini API usage only.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}    

[tomy-kyu]

I see. Understood.
I thought it would be better to at least revise the explanation. After updating, the authentication failed immediately and returned an error.

Also, the intent is unclear, so I wish that part could be clarified. It would be helpful to include a description explaining why this approach is taken—for instance, that it's done because the Gemini API key issued from there can handle everything comprehensively without needing the Vertex AI API key.