Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature Request] 请求适配阿里云deepseek模型think内容 #6183

Open
smoothan opened this issue Feb 8, 2025 · 19 comments
Open

[Feature Request] 请求适配阿里云deepseek模型think内容 #6183

smoothan opened this issue Feb 8, 2025 · 19 comments
Labels
enhancement New feature or request

Comments

@smoothan
Copy link

smoothan commented Feb 8, 2025

🥰 Feature Description

目前阿里云也提供了deepseek-r1和deepseek-v3模型,其返回结果中的思考内容也包含在中。通过custom_model新增deepseek-r1@Alibabadeepseek-v3@Alibaba 时可以使用,但是不显示think的内容。

🧐 Proposed Solution

请求在页面增加思考内容的显示

📝 Additional Information

No response

@smoothan smoothan added the enhancement New feature or request label Feb 8, 2025
@Issues-translate-bot
Copy link

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


Title: [Feature Request] Request to adapt to think content of Alibaba Cloud deepseek model

@capric98
Copy link

capric98 commented Feb 9, 2025

ByteDance模式的r1模型目前也不显示think内容,但如果将其api endpoint填入DEEPSEEK_URL则可正常显示think,希望也能一并适配;并且ByteDance的模型名称是部署名称,可能需要单独处理一下超时的问题(之前的一个patch通过模型名后缀判断是否使用更长的超时时间)

@Issues-translate-bot
Copy link

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


The r1 model of ByteDance mode does not currently display think content, but if you fill in its API endpoint into DEEPSEEK_URL, you can display think normally, and hope it can also be adapted together; and the model name of ByteDance is the deployment name, and the timeout may need to be handled separately. The problem (a previous patch used the model name suffix to determine whether to use a longer timeout time)

@qwe325207
Copy link

ByteDance模式的r1模型目前也不显示think内容,但如果将其api endpoint填入DEEPSEEK_URL则可正常显示think,希望也能一并适配;并且ByteDance的模型名称是部署名称,可能需要单独处理一下超时的问题(之前的一个patch通过模型后缀判断是否使用更长的超时时间)

我也是遇到这个问题 ,请问这个具体是修改那个文件的代码呢? 谢谢

@Issues-translate-bot
Copy link

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


The r1 model of ByteDance mode does not currently display think content, but if you fill in its API endpoint into DEEPSEEK_URL, you can display think normally, and hope it can also be adapted together; and the model name of ByteDance is the deployment name, which may need to be processed separately. Timeout problem (a previous patch used the model suffix to determine whether to use a longer timeout time)

I also encountered this problem. Is this specific code to modify the file? Thanks

@capric98
Copy link

ByteDance模式的r1模型目前也不显示think内容,但如果将其api endpoint填入DEEPSEEK_URL则可正常显示think,希望也能一并适配;并且ByteDance的模型名称是部署名称,可能需要单独处理一下超时的问题(之前的一个patch通过模型后缀判断是否使用更长的超时时间)

我也是遇到这个问题 ,请问这个具体是修改那个文件的代码呢? 谢谢

c2edfec#diff-9061a6ddb3bf8c4c501a29c7d29290ef866f61b8fdbabae5ec1662a8a26e49c2

@Issues-translate-bot
Copy link

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


The r1 model of ByteDance mode does not currently display think content, but if you fill in its API endpoint into DEEPSEEK_URL, you can display think normally, and hope it can also be adapted together; and the model name of ByteDance is the deployment name, which may need to be processed separately Let’s find out the problem of timeout (a previous patch used the model suffix to determine whether to use a longer timeout time)

I also encountered this problem. Is this specific code to modify the file? Thanks

c2edfec#diff-9061a6ddb3bf8c4c501a29c7d29290e f866f61b8fdbabae5ec1662a8a26e49c2

@Issues-translate-bot
Copy link

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


How was it configured? Boss, I also use Alibaba Cloud's deepseek-r1 api, but I have not successfully configured it in Next chat

@GOUKI9999
Copy link

https://bailian.console.aliyun.com/#/model-market/detail/deepseek-r1?tabKey=sdk
同请求,文档中有明确的展示思考过程的对应接口说明

@Issues-translate-bot
Copy link

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


https://bailian.console.aliyun.com/#/model-market/detail/deepseek-r1?tabKey=sdk
The same request, the document has a clear description of the corresponding interface to show the thinking process.

@xiaobai3310
Copy link

咋配置的啊?大佬,我用的也是阿里云的deepseek-r1 api,但是一直没在Next chat配置成功

搞定,且有think过程,main分支已经支持了的

@Issues-translate-bot
Copy link

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


How to configure it? Boss, I also use Alibaba Cloud's deepseek-r1 api, but I have not successfully configured it in Next chat

It's done, and there is a think process, the main branch has already supported it

@bestsanmao
Copy link
Contributor

#6204 适配 阿里 和 字节 的r1

@Issues-translate-bot
Copy link

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


#6204 adapts to Ali and bytes r1

@theores
Copy link

theores commented Feb 11, 2025

咋配置的啊?大佬,我用的也是阿里云的deepseek-r1 api,但是一直没在Next chat配置成功

搞定,且有think过程,main分支已经支持了的

怎么搞定的啊

@Issues-translate-bot
Copy link

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


How to configure it? Boss, I also use Alibaba Cloud's deepseek-r1 api, but I have not successfully configured it in Next chat

Get it done, and there is a think process, the main branch has already supported it

How did it work

@xiaobai3310
Copy link

咋配置的啊?大佬,我用的也是阿里云的deepseek-r1 api,但是一直没在Next chat配置成功

搞定,且有think过程,main分支已经支持了的

怎么搞定的啊

我用的vercel搭建的,用的是阿里云百炼的api,配置如下:
DEEPSEEK_API_KEY=sk-xxxx
DEEPSEEK_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
CUSTOM_MODELS=+deepseek-r1@DeepSeek,+deepseek-v3@DeepSeek

@Issues-translate-bot
Copy link

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


How to configure it? Boss, I also use Alibaba Cloud's deepseek-r1 api, but I have not successfully configured it in Next chat

Get it done, and there is a think process, the main branch has already supported it

How did it work

I built the vercel with Alibaba Cloud Bailian API, and the configuration is as follows:
DEEPSEEK_API_KEY=sk-xxxx
DEEPSEEK_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
CUSTOM_MODELS=+deepseek-r1@DeepSeek,+deepseek-v3@DeepSeek

@Issues-translate-bot
Copy link

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


How to configure it? Boss, I also use Alibaba Cloud's deepseek-r1 api, but I have not successfully configured it in Next chat

Get it done, and there is a think process, the main branch has already supported it

Which main branch to use

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

8 participants