[Feature Request] 智谱清言的api可以填入环境变量OPENAI_API_KEY吗？

[BiFangKNT]

🥰 需求描述

由于我不需要openai的模型，我可以不填OPENAI_API_KEY，或者把智谱清言（chatglm）的api填入吗？

🧐 解决方案

OPENAI_API_KEY为非必填项，或者可以填入智谱清言的api

📝 补充信息

No response

[Issues-translate-bot]

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

---

Title: [Feature Request] Can the environment variable OPENAI_API_KEY be filled in for Zhipu Qingyan’s API?

🥰 Description of requirements

Can I not fill in OPENAI_API_KEY, or fill in the api of Zhipu Qingyan?

🧐 Solution

OPENAI_API_KEY is optional, or you can fill in the api of Zhipu Qingyan

📝 Supplementary information

No response

[QAbot-zh]

教你一招，BASE_URL 填 https://open.bigmodel.cn/api/paas/v4/chat/completions#，OPENAI_API_KEY 填智谱的 api，自定义模型里加上智谱的那些模型，就可以用了

[Issues-translate-bot]

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

---

Let me teach you a trick, BASE_URL fill in https://open.bigmodel.cn/api/paas/v4/chat/completions#, OPENAI_API_KEY fill in the api of Zhipu, add those models of Zhipu to the custom model, you can use it Got it

[BiFangKNT]

教你一招，BASE_URL 填 https://open.bigmodel.cn/api/paas/v4/chat/completions#，OPENAI_API_KEY 填智谱的 api，自定义模型里加上智谱的那些模型，就可以用了

我上面说的就是这个意思，想问这样行不行🤣
但是这两个环境变量好像只能各填一个吧？我想部署多个模型的话，能填进去多个然后用逗号隔开这样吗？比如同时填glm和deepseek的

[Issues-translate-bot]

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

---

Teach you a trick, BASE_URL fill in https://open.bigmodel.cn/api/paas/v4/chat/completions#, OPENAI_API_KEY fill in the api of Zhipu, add those models of Zhipu to the custom model, and you can Used

This is what I said above, I would like to ask if this is okay🤣
But it seems that only one of these two environment variables can be filled in, right? If I want to deploy multiple models, can I fill in multiple models and separate them with commas?

[QAbot-zh]

那确实不行

[Issues-translate-bot]

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

---

That really doesn't work

[BiFangKNT]

那确实不行

那就难整了，项目官方也没提供这两个模型的环境变量。
不过之前刷到过一个教程说，好像可以另外部署个叫one api的项目，似乎可以整合多个模型，然后输出单个url和密钥，填入nextchat这个项目的这两个环境变量中，不知道行不行

[Issues-translate-bot]

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

---

That really doesn't work

That would be difficult. The project official did not provide environment variables for these two models.
However, I came across a tutorial before that said that it seems that you can deploy another project called one api. It seems that you can integrate multiple models, and then output a single URL and key, and fill in the two environment variables of the nextchat project. I don’t know. OK or not

[QAbot-zh]

那是可以的，我就在用，好多类型都可以塞里面

[Issues-translate-bot]

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

---

That's okay, I'm using it, many types can be stuffed in it

[BiFangKNT]

那是可以的，我就在用，好多类型都可以塞里面

我填了那两个环境变量进去发现没有我的模型，是不是要填一下那个CUSTOM_MODELS环境变量来自定义模型名？

[Issues-translate-bot]

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

---

That's okay, I'm using it, many types can be stuffed in it

I filled in the two environment variables and found that my model was not there. Do I need to fill in the CUSTOM_MODELS environment variable to customize the model name?

[BiFangKNT]

那是可以的，我就在用，好多类型都可以塞里面

另外我想问一下glm-4-alltools在这个项目里怎么用？好像要先设置tools，但是项目没有这个设置

[Issues-translate-bot]

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

---

That's okay, I'm using it, many types can be stuffed in it

In addition, I would like to ask how glm-4-alltools is used in this project? It seems that tools need to be set up first, but the project does not have this setting.

[BiFangKNT]

那是可以的，我就在用，好多类型都可以塞里面

而且这个项目好像不支持上传文件和联网，我是用docker部署的，有没有好的办法？好像可以提供插件支持，但是不太清楚怎么搞

[Issues-translate-bot]

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

---

That's okay, I'm using it, many types can be stuffed in it

Moreover, this project does not seem to support uploading files and networking. I deployed it using docker. Is there any good way? It seems that plug-in support can be provided, but I’m not sure how to do it.

[Dogtiti]

最新版本已经支持插件啦
#5353

[Issues-translate-bot]

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

---

The latest version already supports plug-ins