Results 30 comments of shelltea

I encountered the same issue, and it was resolved by setting the base model to `gpt-4`. However, this setup results in the inability to use the `vision` feature. I hope...

In version 0.6.7, this issue has been fixed and can now be closed.

The same issue, still unable to use Gemini. Not sure if these codes have really been tested?

This issue was resolved through the following methods: Modify this file [llm.py](https://github.com/langgenius/dify/blob/main/api/core/model_runtime/model_providers/google/llm/llm.py#L204) ``` new_client_manager.configure(api_key=credentials["google_api_key"], transport='rest') ```

@AlbumenJ 你好,请问3.0 的 alpha 预览版本预计什么时候会发布?

Encountered the same issue, it did not exist in version 0.6.11, but appeared after the upgrade.

@crazywoola @laipz8200 Hello, could you please fix this issue as soon as possible? Because all the models I use are gpt4o on Azure, and now this problem is causing all...

> Hi @shelltea! We are still tracking the issue. Is your model version also `2024-05-13`?