[Bug]

[qi-mooo]

📦 部署方式

Docker

📌 软件版本

275308b

💻 系统环境

Other Linux

📌 系统版本

arch

🌐 浏览器

Chrome

📌 浏览器版本

120

🐛 问题描述

USE_OPENAI_ENDPOINT_FOR_ALL_MODELS
功能失效，即使将他设置成 1 ，在实际请求时其他模型还是使用模型路由，并没有成功切换成 OpenAI 路由

📷 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]

[MuRo-J]

还有一个附近的问题,不过比题主这个问题要轻微,上游的动静实在太大,DEFAULT_MODEL参数也失效了,但是他们后续补丁处理了,得辛苦大佬需要的时候合并

btw目前已知的一个常用环境变量设置方式:
CUSTOM_MODELS
举例:
原使用方法:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

会出现一个问题就是,gpt-4o会同时命中openai + azure等,会导致模型列表多处很多可能不需要使用的模型列表
改为用法:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

[Issues-translate-bot]

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

---

There is another nearby problem, but it is less severe than the one in question. The upstream movement was too big, and the DEFAULT_MODEL parameter also failed. However, they solved it with subsequent patches. It’s a big headache. They are still working hard.

[Hk-Gosuto]

前些天的代码合并搞出了好多问题 晚一点我处理一下

[Issues-translate-bot]

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

---

The code merge a few days ago caused a lot of problems. I'll deal with them later.

[Hk-Gosuto]

还有一个附近的问题,不过比题主这个问题要轻微,上游的动静实在太大,DEFAULT_MODEL参数也失效了,但是他们后续补丁处理了,得辛苦大佬需要的时候合并

btw目前已知的一个常用环境变量设置方式: CUSTOM_MODELS 举例: 原使用方法:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

会出现一个问题就是,gpt-4o会同时命中openai + azure等,会导致模型列表多处很多可能不需要使用的模型列表 改为用法:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

没找到上游这部分的代码的差别 可否指出修复部分的代码 我单独合一下这部分

[Issues-translate-bot]

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

---

There is another nearby problem, but it is less severe than the one in question. The upstream movement is too big, and the DEFAULT_MODEL parameter is also invalid. However, they have solved it in subsequent patches. Thank you for your hard work and merge it when needed.

btw A commonly used environment variable setting method currently known: CUSTOM_MODELS Example: Original usage:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude -3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3- 5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

One problem that will arise is that gpt-4o will hit openai + azure, etc. at the same time, which will result in many model lists that may not be needed. Change the usage:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5- flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5 -sonnet-20240620@anthropic=claude-3.5-sonnet

I couldn't find the difference between this part of the upstream code. Can you point out the code to fix it? I'll combine this part separately.

[MuRo-J]

还有一个附近的问题,不过比题主这个问题要轻微,上游的动静实在太大,DEFAULT_MODEL参数也失效了,但是他们后续补丁处理了,得辛苦大佬需要的时候合并
btw目前已知的一个常用环境变量设置方式: CUSTOM_MODELS 举例: 原使用方法:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

会出现一个问题就是,gpt-4o会同时命中openai + azure等,会导致模型列表多处很多可能不需要使用的模型列表 改为用法:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

没找到上游这部分的代码的差别 可否指出修复部分的代码 我单独合一下这部分

感谢您费心,
DEFAULT_MODULE他们通过
ChatGPTNextWeb@2fd68bc
修复了

CUSTUM_MODULE暂时无解,只能自己修改变量用法解决同时有 chatgpt-4o(OpenAI) 和chatgpt-4o(Azure)的问题
如这里
ChatGPTNextWeb#5030

[Issues-translate-bot]

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

---

There is also a nearby problem, but it is less severe than the problem of the subject. The upstream movement is too big, and the DEFAULT_MODEL parameter is also invalid. However, they have solved it in subsequent patches. Thank you for your hard work and merge it when needed.
btw A commonly used environment variable setting method currently known: CUSTOM_MODELS Example: Original usage:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+ claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3 -5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

A problem will arise is that gpt-4o will hit openai + azure, etc. at the same time, which will result in many model lists that may not be used. Change the usage:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5 -flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3- 5-sonnet-20240620@anthropic=claude-3.5-sonnet

I couldn’t find the difference between this part of the upstream code. Can you point out the code to fix it? I’ll combine this part separately.

Thank you for your trouble,
DEFAULT_MODULE they pass
ChatGPTNextWeb@2fd68bc
fixed

CUSTUM_MODULE has no solution for the time being. You can only modify the variable usage yourself to solve the problem of chatgpt-4o (OpenAI) and chatgpt-4o (Azure) at the same time.
Like here
ChatGPTNextWeb#5030

[Hk-Gosuto]

还有一个附近的问题,不过比题主这个问题要轻微,上游的动静实在太大,DEFAULT_MODEL参数也失效了,但是他们后续补丁处理了,得辛苦大佬需要的时候合并
btw目前已知的一个常用环境变量设置方式: CUSTOM_MODELS 举例: 原使用方法:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

会出现一个问题就是,gpt-4o会同时命中openai + azure等,会导致模型列表多处很多可能不需要使用的模型列表 改为用法:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

没找到上游这部分的代码的差别 可否指出修复部分的代码 我单独合一下这部分

感谢您费心, DEFAULT_MODULE他们通过 ChatGPTNextWeb@2fd68bc 修复了

CUSTUM_MODULE暂时无解,只能自己修改变量用法解决同时有 chatgpt-4o(OpenAI) 和chatgpt-4o(Azure)的问题 如这里 ChatGPTNextWeb#5030

单独 merge 了 collectModelTableWithDefaultModel 方法，可以试试看问题是否解决。

[Issues-translate-bot]

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

---

There is also a nearby problem, but it is less severe than the problem of the subject. The upstream movement is too big, and the DEFAULT_MODEL parameter is also invalid. However, they have solved it in subsequent patches. Thank you for your hard work and merge it when needed.
btw A commonly used environment variable setting method currently known: CUSTOM_MODELS Example: Original usage:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo, +claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude- 3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

A problem will arise is that gpt-4o will hit openai + azure, etc. at the same time, which will result in many model lists that may not be used. Change the usage:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini- 1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3 -5-sonnet-20240620@anthropic=claude-3.5-sonnet

I couldn’t find the difference between this part of the upstream code. Can you point out the code to fix it? I’ll combine this part separately.

Thanks for your trouble, DEFAULT_MODULE they were fixed by ChatGPTNextWeb@2fd68bc

CUSTUM_MODULE has no solution for the time being. You can only modify the variable usage yourself to solve the problem of chatgpt-4o (OpenAI) and chatgpt-4o (Azure) at the same time, as shown here [ChatGPTNextWeb#5030](https://github.com/ChatGPTNextWeb/ChatGPT- Next-Web/issues/5030) 

You can merge the collectModelTableWithDefaultModel method separately and try to see if the problem is solved.

[MuRo-J]

还有一个附近的问题,不过比题主这个问题要轻微,上游的动静实在太大,DEFAULT_MODEL参数也失效了,但是他们后续补丁处理了,得辛苦大佬需要的时候合并
btw目前已知的一个常用环境变量设置方式: CUSTOM_MODELS 举例: 原使用方法:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude-3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

会出现一个问题就是,gpt-4o会同时命中openai + azure等,会导致模型列表多处很多可能不需要使用的模型列表 改为用法:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini-1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude-3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

没找到上游这部分的代码的差别 可否指出修复部分的代码 我单独合一下这部分

感谢您费心, DEFAULT_MODULE他们通过 ChatGPTNextWeb@2fd68bc 修复了
CUSTUM_MODULE暂时无解,只能自己修改变量用法解决同时有 chatgpt-4o(OpenAI) 和chatgpt-4o(Azure)的问题 如这里 ChatGPTNextWeb#5030

单独 merge 了 collectModelTableWithDefaultModel 方法，可以试试看问题是否解决。

再次说声辛苦啦,DEFAUT_MODULE正常了,可以正常起作用

[Issues-translate-bot]

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

---

There is another nearby problem, but it is less serious than the problem of the subject. The upstream movement is too big, and the DEFAULT_MODEL parameter is also invalid. However, they have solved it in subsequent patches. Thank you for your hard work and merge it when needed.
btw A commonly used environment variable setting method currently known: CUSTOM_MODELS Example: Original usage:

-all,+gpt-4o,+gpt-4o-mini,+gpt-4-turbo,+gemini-1.5-flash-latest,+gemini-1.5-pro-latest,+gpt-3.5-turbo ,+claude-3-opus-20240229,+claude-3-5-sonnet-20240620,gemini-1.5-pro-latest=gemini-1.5-pro,claude-3-opus-20240229=claude-3-opus,claude -3-5-sonnet-20240620=claude-3.5-sonnet,gemini-1.5-flash-latest=gemini-1.5-flash

A problem will arise is that gpt-4o will hit openai + azure, etc. at the same time, which will result in many model lists that may not be used. Change the usage:

-all,+gpt-4o@openai=gpt-4o,+gpt-4o-mini@openai=gpt-4o-mini,+gpt-4-turbo@openai=gpt-4-turbo,+gemini -1.5-flash-latest@google=gemini-1.5,+gemini-1.5-pro-latest@google=gemini-1.5-pro,+claude-3-opus-20240229@anthropic=claude-3-opus,+claude- 3-5-sonnet-20240620@anthropic=claude-3.5-sonnet

I couldn’t find the difference between this part of the upstream code. Can you point out the code to fix it? I’ll combine this part separately.

Thanks for your trouble, DEFAULT_MODULE they fixed via ChatGPTNextWeb@2fd68bc
CUSTUM_MODULE has no solution for the time being. You can only modify the variable usage yourself to solve the problem of chatgpt-4o (OpenAI) and chatgpt-4o (Azure) at the same time, as shown here [ChatGPTNextWeb#5030](https://github.com/ChatGPTNextWeb/ChatGPT -Next-Web/issues/5030) 

Merge the collectModelTableWithDefaultModel method separately. You can try it to see if the problem is solved.

Thank you again for your hard work. DEFAUT_MODULE is now normal and can work normally.