NextChat
NextChat copied to clipboard
ChatGPTNextWeb
[Bug] window版本和web版本都能用gemini，只有docker版本2.12.2的用不了，提示{ "error": true, "message": "Request failed" } 怎么解决
Bug Description
window版本和web版本都能用gemini，只有docker版本2.12.2的用不了，提示{ "error": true, "message": "Request failed" }
怎么解决
Steps to Reproduce
提示{ "error": true, "message": "Request failed" }
Expected Behavior
提示{ "error": true, "message": "Request failed" }
Screenshots
提示{ "error": true, "message": "Request failed" }
Deployment Method
- [X] 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
提示{ "error": true, "message": "Request failed" }
Bot detected the issue body's language is not English, translate it automatically.
Are there any shell commands used during deployment, including error screenshots?
请不要使用one-api代理后的gemini-pro 因为one-api会转换成OpenAI格式 如果必须使用one-api 请自定义模型映射代码 除此之外 如果直接填写Google API Key出现问题 说明调用的接口不对 现在谷歌最新的接口是/v1beta而不是/v1
Bot detected the issue body's language is not English, translate it automatically.
Please do not use gemini-pro after one-api proxy because one-api will convert it to OpenAI format If you must use one-api, please customize the model mapping code In addition, if there is a problem directly filling in the Google API Key, it means that the calling interface is wrong. Now Google’s latest interface is /v1beta instead of /v1.