ChatGPTNextWeb
[Bug] 无法使用豆包大模型
📦 部署方式
官方安装包
📌 软件版本
v2.13.1
💻 系统环境
Windows
📌 系统版本
22H2
🌐 浏览器
Chrome
📌 浏览器版本
126.0.6478.128
🐛 问题描述
豆包的接口地址是 https://ark.cn-beijing.volces.com/api/v3 ，但填写了没用，始终请求 https://ark.cn-beijing.volces.com/api/v1/chat/completions ，没有 v3
📷 复现步骤
桌面端不知道哪里有开发者工具，拿网页演示一下：
https://github.com/user-attachments/assets/edae0d06-3daa-4261-828b-7dfd42468d85
🚦 期望结果
正常请求正常使用
📝 补充信息
No response
Bot detected the issue body's language is not English, translate it automatically.
Title: [Bug] Unable to use bean bag large model
使用自定义接口的时候，接口地址填写： https://ark.cn-beijing.volces.com
始终请求 https://ark.cn-beijing.volces.com/api/v1/chat/completions ，没有 v3 说明模型被当成了OpenAI的接口使用。
豆包的使用需要配合CUSTOM_MODELS一起(配置部署名称)使用
填写 https://ark.cn-beijing.volces.com/ 之后请求的地址还是会加上 v1/chat/completions
豆包的使用需要配合CUSTOM_MODELS一起(配置部署名称)使用
这是指什么？不是接入点名称吗？
Bot detected the issue body's language is not English, translate it automatically.
I'm still not successful here. I don't know where else I have the problem.
Bot detected the issue body's language is not English, translate it automatically.
Just like me, I don’t know what’s going on
对话框跟我情况一样，像是逻辑有bug，抓了下包，刚开始会通过OPTIONS去访问https://ark.cn-beijing.volces.com/api/v3/chat/completions，但是没有带任何auth key，报了401.
而且这个行为也很奇怪，本地试了下及时带了key，也会报404。交互协议对不上
@Crazykev OPTIONS请求是HTTP协议里面在跨域的时候的标准做法。先通过OPTIONS接口检查是否支持跨域。检查通过之后再发送POST请求。
Bot detected the issue body's language is not English, translate it automatically.
@Crazykev OPTIONS request is a standard practice in the HTTP protocol when making cross-domain requests. First check whether cross-domain is supported through the OPTIONS interface. Send the POST request after the check passes.
{ "error": true, "message": "you are not allowed to use Doubao-vision-pro-32k model" } 这咋回事啊？
我也遇到相同的问题，但是排查了配置信息，也没发现哪里出错.截图里面的key填写endpoint只是尝试，实际换成key也不管用
{ "error": true, "message": "NetworkError when attempting to fetch resource." }
any updates on this? Thank you reply Already solve it, I was just only modify docker config.
services:
chatgpt-next-web:
image: yidadaa/chatgpt-next-web:latest
restart: unless-stopped
ports:
- "127.0.0.1:3000:3000"
environment:
- CUSTOM_MODELS=-all,+Doubao-pro-128k@bytedance=ep-2025010908xxxx-xxxxx
- BYTEDANCE_API_KEY=xxxxx
any updates on this? Thank you reply Already solve it, I was just only modify docker config.
services: chatgpt-next-web: image: yidadaa/chatgpt-next-web:latest restart: unless-stopped ports: - "127.0.0.1:3000:3000" environment: - CUSTOM_MODELS=-all,+Doubao-pro-128k@bytedance=ep-2025010908xxxx-xxxxx - BYTEDANCE_API_KEY=xxxxx
Hi, have you tried vercel deploy?
接口地址：https://ark.cn-beijing.volces.com
密钥：
BYTEDANCE_API_KEY=xxxxx CUSTOM_MODELS=-all,+Doubao-1.5-pro-32k@bytedance=ep-xxxxxxx DEFAULT_MODEL=Doubao-1.5-pro-32k
Bot detected the issue body's language is not English, translate it automatically.
As mentioned in Issue #6177, changing the bytedance in the custom model name to ByteDance has an effect