ChatGPTNextWeb
[Bug] 自定义模型和预设模型名称相同时，会被预设模型覆盖
Bug Description
我使用了第三方中转API，其给出的可用模型名称有很多与预设模型名称相同。这家的中转方式是，无论是谁家的模型，一律使用OpenAI API。因此，当我想要调用claude-3-opus-20240229时，不能使用预设的opus，只能输入自定义模型名。但当我在设置的自定义模型名输入框输入精确的claude-3-opus-20240229时，此选项不会出现。当在名称内随便加入或删除字母，对应选项出现。因此判断为自定义模型名被预设模型覆盖，这种情况下我无法使用这家的服务
Steps to Reproduce
- 设置-自定义模型名
- 输入 claude-3-haiku-20240307,claude-3-opus-20240229,claude-3-sonnet-20240229
- 展开模型下拉框，可以看到没有新模型
- 自定义模型名改为claude-3-haiku-2024030777,claude-3-opus-2024022999,claude-3-sonnet-2024022999
- 模型下拉框中，对应的模型名出现
Expected Behavior
输入名称与预设一致的自定义模型名时，其应当另外展示出来。另外也许应该提供自定义模型名与模型服务商自由绑定的能力
Screenshots
No response
Deployment Method
- [ ] Docker
- [ ] Vercel
- [ ] Server
Desktop OS
No response
Desktop Browser
No response
Desktop Browser Version
No response
Smartphone Device
No response
Smartphone OS
No response
Smartphone Browser
No response
Smartphone Browser Version
No response
Additional Logs
No response
Bot detected the issue body's language is not English, translate it automatically.
Title: [Bug] When the custom model and the preset model have the same name, they will be overwritten by the preset model.
Bot detected the issue body's language is not English, translate it automatically.
The model name is bound to the model provider and model access method. It is recommended to use one-api for model name redirection
模型名与模型提供商和模型访问方式绑定了。建议用one-api做次模型名重定向
是否可以在更改相对较小的情况下让自定义模型名与默认模型名即便重复也可以作为不同的选项出现？如果可以的话，我空了也许可以看看能不能改
Bot detected the issue body's language is not English, translate it automatically.
The model name is bound to the model provider and model access method. It is recommended to use one-api for model name redirection
Is it possible, with relatively minor changes, to have a custom model name appear as different options than the default model name, even if they are duplicates? If possible, I might be able to see if I can change it if I have some free time.
We will introduce new api design in V3 to resolve this issue by implementing Model Manager.
https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/pull/4930#issuecomment-2213755908
后续有计划新增一个自定义CUSTOM_PROVIDER的方式，配合CUSTOM_MODELS 做到可以配置和内置模型名字一样（处理消息，构造请求逻辑一致）但是走自定义的API发送请求的逻辑。
这样就能做到不会覆盖预设的模型。
Bot detected the issue body's language is not English, translate it automatically.
https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web/pull/4930#issuecomment-2213755908
There are plans to add a custom CUSTOM_PROVIDER method in the future. With CUSTOM_MODELS, it can be configured with the same name as the built-in model (processing messages and constructing request logic) but using a custom API to send request logic.
This way the preset model will not be overwritten.
