[Feature Request]: 完善gemini的支持

[rtxneko]

Problem Description

目前使用脚本部署的话是强制要求填写openai key的
使用docker部署的话，填了google aistudio key后默认依然会请求openai的api

Solution Description

额 docker部署时加个能选择的参数？

Alternatives Considered

No response

Additional Context

No response

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Title: [Feature Request]: Improve gemini support

Problem Description

Currently, if you use script deployment, it is mandatory to fill in the openai key.
If you use docker to deploy, after filling in the google aistudio key, the openai api will still be requested by default.

Solution Description

Um, add an optional parameter when deploying docker?

Alternatives Considered

No response

Additional Context

No response

[Kosette]

添加环境变量CUSTOM_MODELS=-all,+gemini-pro,+gemini-pro-vision试试

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Try adding the environment variable CUSTOM_MODELS=-all,+gemini-pro,+gemini-pro-vision

[pp6451303]

window版本和web版本都能用gemini，只有docker版本的用不了，提示{
"error": true,
"message": "Request failed"
}

怎么解决