[Bug] 同时填入openai和gemini的key和代理地址，貌似默认使用openai的key，此时如何切换或使用gemini的key

[DreamofInfinity1]

📦 部署方式

Other

📌 软件版本

v2.15.2

💻 系统环境

Windows

📌 系统版本

win11

🌐 浏览器

Edge

📌 浏览器版本

edge

🐛 问题描述

部署在cloudflare，同时填入openai和gemini的key和代理地址，貌似默认使用openai的key，此时如何切换或使用gemini的key

📷 复现步骤

No response

🚦 期望结果

No response

📝 补充信息

No response

[Issues-translate-bot]

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

---

Title: [Bug] Fill in the keys and proxy addresses of openai and gemini at the same time. It seems that the openai key is used by default. How to switch or use the gemini key at this time?

[H0llyW00dzZ]

📦 部署方式

Other

📌 软件版本

v2.15.2

💻 系统环境

Windows

📌 系统版本

win11

🌐 浏览器

Edge

📌 浏览器版本

edge

🐛 问题描述

部署在cloudflare，同时填入openai和gemini的key和代理地址，貌似默认使用openai的key，此时如何切换或使用gemini的key

📷 复现步骤

No response

🚦 期望结果

No response

📝 补充信息

No response

You have to go to the settings, I think. For example, like this:

Then, go back to the chat because the old desktop version, which is built into binaries (client-side only), still works for me using this method.

[DreamofInfinity1]

@H0llyW00dzZ 我是将openai和gemini的key和代理地址以环境变量嵌入程序，我想在此种情景下进行切换

[Issues-translate-bot]

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

---

@H0llyW00dzZ I embed the keys and proxy addresses of openai and gemini into the program as environment variables. I want to switch in this situation.

[lloydzhou]

[Issues-translate-bot]

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

---

[DreamofInfinity1]

@lloydzhou 您这样貌似仅仅用来区别不同厂商相同模型名的情况。我的意思指有两个api，一个是别人的new-api，里面虽然包含gemini的各种模型，但实际无法使用。第二个api是自己Geminipro的api，可以使用。我想问的是两个api中模型列表重叠，一个不可用，如何调用Google api来使用gemini

[Issues-translate-bot]

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

---

@lloydzhou You seem to be just using it to distinguish between different manufacturers with the same model name. What I mean is that there are two APIs. One is someone else's new-API. Although it contains various models of gemini, it cannot actually be used. The second api is your own Geminipro api and can be used. What I want to ask is that the model lists in the two APIs overlap and one is unavailable. How to call the Google API to use gemini?