AI 助手全套开源解决方案，自带运营管理后台，开箱即用。集成了 ChatGPT, Azure, ChatGLM,讯飞星火，文心一言等多个平台的大语言模型。支持 MJ AI 绘画，Stable Diffusion AI 绘画，微博热搜等插件工具。采用 Go + Vue3 + element-plus 实现。
Home Page: https://ai.r9it.com
License: MIT License
给可以每个用户设置GPT 模型范围，方便对用户权限进行管理。
高级用户：gpt-4-Mobile, gpt-4, gpt-3.5
普通用户：gpt-3.5
一部分高级用户享用Gpt4 普通用户就用gpt3.5
控制成本
GPT-3.5
Windows
请看图，在admin面板中可以增加用户，用户自行注册就不行了
No response
把你的配置信息粘贴到这里GPT-3.5
Linux
启动容器报错，看倒数第三行。
(base) pi@pi:~/Desktop/GPT/chatgpt-plus/docker$ docker-compose up
Starting chatgpt-plus-vue ... done
Starting chatgpt-plus-go ... done
Attaching to chatgpt-plus-go, chatgpt-plus-vue
chatgpt-plus-go | 2023-08-21T23:46:10.444+0800 INFO api/main.go:61 Loading config file: config.toml
chatgpt-plus-go | [Fx] PROVIDE *types.AppConfig <= main.main.func2()
chatgpt-plus-vue | /docker-entrypoint.sh: /docker-entrypoint.d/ is not empty, will attempt to perform configuration
chatgpt-plus-vue | /docker-entrypoint.sh: Looking for shell scripts in /docker-entrypoint.d/
chatgpt-plus-vue | /docker-entrypoint.sh: Launching /docker-entrypoint.d/10-listen-on-ipv6-by-default.sh
chatgpt-plus-go | [Fx] PROVIDE *core.AppServer <= chatplus/core.NewServer()
chatgpt-plus-vue | 10-listen-on-ipv6-by-default.sh: info: /etc/nginx/conf.d/default.conf is not a file or does not exist
chatgpt-plus-go | [Fx] PROVIDE *gorm.Config <= chatplus/store.NewGormConfig()
chatgpt-plus-vue | /docker-entrypoint.sh: Launching /docker-entrypoint.d/20-envsubst-on-templates.sh
chatgpt-plus-vue | /docker-entrypoint.sh: Launching /docker-entrypoint.d/30-tune-worker-processes.sh
chatgpt-plus-go | [Fx] PROVIDE *gorm.DB <= chatplus/store.NewMysql()
chatgpt-plus-vue | /docker-entrypoint.sh: Configuration complete; ready for start up
chatgpt-plus-go | [Fx] PROVIDE *store.LevelDB <= chatplus/store.NewLevelDB()
chatgpt-plus-go | [Fx] PROVIDE *xdb.Searcher <= main.main.func4()
chatgpt-plus-go | [Fx] PROVIDE *wexin.WeChatBot <= chatplus/modules/wexin.NewWeChatBot()
chatgpt-plus-go | [Fx] PROVIDE function.FuncZaoBao <= main.main.func6()
chatgpt-plus-go | [Fx] PROVIDE function.FuncWeiboHot <= main.main.func7()
chatgpt-plus-go | [Fx] PROVIDE function.FuncHeadlines <= main.main.func8()
chatgpt-plus-go | [Fx] PROVIDE *handler.ChatRoleHandler <= chatplus/handler.NewChatRoleHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.UserHandler <= chatplus/handler.NewUserHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.ChatHandler <= chatplus/handler.NewChatHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.UploadHandler <= chatplus/handler.NewUploadHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.SmsHandler <= chatplus/handler.NewSmsHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.RewardHandler <= chatplus/handler.NewRewardHandler()
chatgpt-plus-go | [Fx] PROVIDE *handler.CaptchaHandler <= chatplus/handler.NewCaptchaHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.ConfigHandler <= chatplus/handler/admin.NewConfigHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.ManagerHandler <= chatplus/handler/admin.NewAdminHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.ApiKeyHandler <= chatplus/handler/admin.NewApiKeyHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.UserHandler <= chatplus/handler/admin.NewUserHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.ChatRoleHandler <= chatplus/handler/admin.NewChatRoleHandler()
chatgpt-plus-go | [Fx] PROVIDE *admin.RewardHandler <= chatplus/handler/admin.NewRewardHandler()
chatgpt-plus-go | [Fx] PROVIDE *service.AliYunSmsService <= chatplus/service.NewAliYunSmsService()
chatgpt-plus-go | [Fx] PROVIDE *service.CaptchaService <= main.main.func9()
chatgpt-plus-go | [Fx] PROVIDE fx.Lifecycle <= go.uber.org/fx.New.func1()
chatgpt-plus-go | [Fx] PROVIDE fx.Shutdowner <= go.uber.org/fx.(*App).shutdowner-fm()
chatgpt-plus-go | [Fx] PROVIDE fx.DotGraph <= go.uber.org/fx.(*App).dotGraph-fm()
chatgpt-plus-go | [Fx] INVOKE main.main.func3()
chatgpt-plus-go | 2023-08-21T23:46:10.445+0800 INFO core/app_server.go:63 Enabled debug mode
chatgpt-plus-go | 2023-08-21T23:46:10.445+0800 INFO core/app_server.go:126 Session driver: cookie
chatgpt-plus-go | [Fx] INVOKE main.main.func5()
chatgpt-plus-go |
chatgpt-plus-go | 2023/08/21 23:46:10 /go/chatgpt-plus/api/store/mysql.go:23
chatgpt-plus-go | [error] failed to initialize database, got error dial tcp 192.168.1.102:3307: connect: connection refused
chatgpt-plus-go | [Fx] INVOKE main.main.func10()
chatgpt-plus-go | [Fx] INVOKE main.main.func11()
cd docker/mysql
# 创建 mysql 容器
docker-compose up -d
# 导入数据库
docker exec -i chatgpt-plus-mysql sh -c 'exec mysql -uroot -p12345678' < ../../database/chatgpt_plus.sql修改docker/conf/config.toml
ProxyURL = "http://127.0.0.1:7890"
MysqlDns = "root:12345678@tcp(192.168.1.102:3307)/chatgpt_plus?charset=utf8mb4&collation=utf8mb4_unicode_ci&parseTime=True&loc=Local"修改docker/conf/nginx/conf.d/chatgpt-plus.conf
proxy_pass http://192.168.1.102:5678; # 这里改成后端服务的内网 IP 地址
}
# 静态资源转发
location /static/ {
proxy_pass http://192.168.1.102:5678;梯子的端口
ip地址
下面那个ip，gpt说是无线网络的，不用管它
GPT-3.5
Linux
更新到3.0.6.3后登录后台，输入正确的账号密码后不跳转到仪表盘；还有就是关于导入数据库文件的问题，我应该导入哪个sql文件？您目前/database目录下有两个sql文件，一个update和一个chatplus的文件，导入这两个有什么区别吗？感谢您的解答
这是输入正确的账号密码后，点击登录没反应
这是输入错误的账号密码后的界面
Is it possible to use caddy, instead of Nginx, because caddy can automatically generate https and it is easier to use the reverse proxy mode
能否使用caddy，替换掉Nginx，因为caddy能够自动生成https，而且更容易使用反代模式
如果使用caddy服务器，就不用单独申请域名的https
GPT-3.5
Linux
二进制包在服务器上运行时，前端绑死的访问localhost啊，这个不能把它提取出来吗
下载了二进制包，在服务器上启动，接口依然请求的localhost
Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "0.0.0.0:5678"
ProxyURL = []
AccessKey = "bzyn7ttujhte6ajpvvsp5sphzaiexi9fjnkupb8san"
[Session]
SecretKey = "1x9v9udkcqnnz8d37v1ub3tksaffbehyafasww82ekagrawjpchseh0jc339fdsm"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = ""
MaxAge = 86400
Secure = true
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = ""
WechatGroup = ""
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 0.9
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
ApiKeys = []```支撑上传文档，基于文档+llm进行问答，例如langchain
langchain
这样可以方便一些用户的问答，
GPT-3.5
Linux
参考了
等配置，均无法使用公网IP访问，报错超时无法访问。
在阿里云服务器上curl http://localhost:8080/chat有结果，查看容器日志也都启动成功，但换成http://公网ip:8080/chat时则访问超时，且已确认将服务器8080端口对外开放。
想请教一下仅修改配置是无法生效嘛，是需要修改哪个地方代码，重新编译后才能用公网ip访问嘛。
把你的配置信息粘贴到这里使用容器版本v3.1.4：
chatgpt-plus.conf配置如下：
map $http_upgrade $connection_upgrade {
default upgrade;
'websocket' upgrade;
}
server {
listen 8080;
server_name 公网ip;
access_log /var/log/access.log;
error_log /var/log/error.log;
index index.html;
root /var/www/app/dist;
location / {
try_files $uri $uri/ /index.html;
location /api/ {
proxy_http_version 1.1;
proxy_connect_timeout 300s;
proxy_read_timeout 300s;
proxy_send_timeout 12s;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_pass http://公网ip:5678; # 这里改成后端服务的内网 IP 地址
}
location /static/ {
proxy_pass http://公网ip:5678; # 这里改成后端服务的内网 IP 地址
}
}
}
config.toml配置如下：
Listen = "0.0.0.0:5678"
ProxyURL = "http://公网ip:7777" # 如 http://127.0.0.1:7777
MysqlDns = "root:12345678@tcp(公网ip:3307)/chatgpt_plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static" # 静态资源的目录
StaticUrl = "/static" # 静态资源访问 URL
AesEncryptKey = ""
WeChatBot = false
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80" # 注意：这个是 JWT Token 授权密钥，生产环境请务必更换
MaxAge = 86400
Domain = "公网ip"
Secure = false
HttpOnly = false
[Manager]
Username = "admin"
Password = "admin123" # 如果是生产环境的话，这里管理员的密码记得修改
[Redis] # redis 配置信息
Host = "公网ip"
Port = 6379
Password = ""
DB = 0
.env.development配置：
VUE_APP_API_HOST=http://公网ip:5678
VUE_APP_WS_HOST=ws://公网ip:5678
VUE_APP_USER=18575670125
VUE_APP_PASS=12345678
VUE_APP_ADMIN_USER=admin
VUE_APP_ADMIN_PASS=admin123
VUE_APP_KEY_PREFIX=ChatPLUS_DEV_
VUE_APP_TITLE="ChatGPT-PLUS V3"
.env.production配置：
VUE_APP_API_HOST=http://公网ip:5678
VUE_APP_WS_HOST=http://公网ip:5678
VUE_APP_KEY_PREFIX=ChatPLUS_
VUE_APP_TITLE="ChatGPT-PLUS V3"
GPT-3.5
Windows
后端路径：“http://localhost:8888/admin/login
前端路径：http://localhost:8888/login
看了之前有人提过的问题，说前端路径应该是 http://localhost:8888/chat 访问这个直接跳转到了 http://localhost:8888/login
不知道是我哪里启动的问题
本地数据库 ，本地 go语言环境； 开发工具 vscode
No response
Listen = "0.0.0.0:5678"
ProxyURL = ""
MysqlDns = "root:123456aA@tcp(127.0.0.1:3306)/chatgpt_plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static"
StaticUrl = "http://localhost/5678/static"
[Session]
Driver = "cookie"
SecretKey = "m0utpsyxjkl5ekgvyiqzmxnrvbdfkv781x6d2baldr56kla8njfa2mtpjocomo0y"
Name = "CHAT_PLUS_SESSION"
Path = "/"
Domain = ""
MaxAge = 86400
Secure = true
HttpOnly = false
SameSite = 2
[Manager]
Username = "admin"
Password = "admin123"
.env.production 配置信息
VUE_APP_API_HOST=http://localhost:5678
VUE_APP_WS_HOST=http://localhost:5678
VUE_APP_BASE_URL=
Windows对make的支持有问题, 网上提到通过chocolate来安装make的方式试了下，依然不OK。
最后摸索了种方案，可以在chatgpt-plus\api\go 目录下，powershell执行如下命令，可以实现打包编译的效果
# GOOS为目标环境的系统，GOARCH为目标环境的CPU架构
$env:CGO_ENABLED=0; $env:GOOS="linux" ; $env:GOARCH="amd64"; go build -o bin/chatgpt-v3-amd64-linux main.go
添加OpenAI的APIkey时，可以自行指定填写接口地址（若不填写则为默认接口地址）
No response
相较于在服务器中使用代理，使用CF或者其他人提供的代理地址对于国内服务器来说更加简单易用，默认情况下可以为官方的接口地址。如果有需求则可以用户自己指定API接口的代理地址。
同时这样做还有一个好处就是目前许多商家提供中转API服务，都需要填写商家指定的接口地址，这些中转API服务提供了许多OpenAI模型可供选择（而且4.0API调用价格比官方低），这样可以节约成本。
GPT-3.5
Linux
运行 go run main.go报错如下：
[Fx] ERROR fx.Invoke(main.main.func3()) called from:
main.main
/root/chatgpt-plus/api/main.go:88
runtime.main
/usr/lib/go-1.20/src/runtime/proc.go:250
Failed: could not build arguments for function "main".main.func3
/root/chatgpt-plus/api/main.go:88:
failed to build *core.AppServer:
could not build arguments for function "chatplus/core".NewServer
/root/chatgpt-plus/api/core/app_server.go:39:
failed to build map[string]fun.Function:
could not build arguments for function "chatplus/service/fun".NewFunctions
/root/chatgpt-plus/api/service/fun/function.go:32:
failed to build *mj.Service:
could not build arguments for function "chatplus/service/mj".NewService
/root/chatgpt-plus/api/service/mj/service.go:33:
failed to build *redis.Client:
received non-nil error from function "chatplus/store".NewRedisClient
/root/chatgpt-plus/api/store/redis.go:10:
read tcp [::1]:42976->[::1]:6379: read: connection reset by peer
[Fx] ERROR Failed to start: could not build arguments for function "main".main.func3
/root/chatgpt-plus/api/main.go:88:
failed to build *core.AppServer:
could not build arguments for function "chatplus/core".NewServer
/root/chatgpt-plus/api/core/app_server.go:39:
failed to build map[string]fun.Function:
could not build arguments for function "chatplus/service/fun".NewFunctions
/root/chatgpt-plus/api/service/fun/function.go:32:
failed to build *mj.Service:
could not build arguments for function "chatplus/service/mj".NewService
/root/chatgpt-plus/api/service/mj/service.go:33:
failed to build *redis.Client:
received non-nil error from function "chatplus/store".NewRedisClient
/root/chatgpt-plus/api/store/redis.go:10:
read tcp [::1]:42976->[::1]:6379: read: connection reset by peer
2023/10/11 11:51:19 could not build arguments for function "main".main.func3 (/root/chatgpt-plus/api/main.go:88): failed to build *core.AppServer: could not build arguments for function "chatplus/core".NewServer (/root/chatgpt-plus/api/core/app_server.go:39): failed to build map[string]fun.Function: could not build arguments for function "chatplus/service/fun".NewFunctions (/root/chatgpt-plus/api/service/fun/function.go:32): failed to build *mj.Service: could not build arguments for function "chatplus/service/mj".NewService (/root/chatgpt-plus/api/service/mj/service.go:33): failed to build *redis.Client: received non-nil error from function "chatplus/store".NewRedisClient (/root/chatgpt-plus/api/store/redis.go:10): read tcp [::1]:42976->[::1]:6379: read: connection reset by peer
exit status 1
No response
把你的配置信息粘贴到这里Listen = "0.0.0.0:5678"
ProxyURL = "" # 如 http://127.0.0.1:7777
MysqlDns = "root:12345678@tcp(mysql:3306)/chatgpt_plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static" # 静态资源的目录
StaticUrl = "/static" # 静态资源访问 URL
AesEncryptKey = ""
WeChatBot = false
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80" # 注意：这个是 JWT Token 授权密钥，生产环境请务必更换
MaxAge = 86400
[Manager]
Username = "admin"
Password = "65132090" # 如果是生产环境的话，这里管理员的密码记得修改
[Redis] # redis 配置信息
Host = "localhost"
Port = 6379
Password = ""
DB = 0
[ApiConfig] # 微博热搜，今日头条等函数服务 API 配置，此为第三方插件服务，如需使用请联系作者开通
ApiURL = ""
AppId = ""
Token = ""
[SmsConfig] # 阿里云短信服务配置
AccessKey = ""
AccessSecret = ""
Product = "Dysmsapi"
Domain = "dysmsapi.aliyuncs.com"
Sign = ""
CodeTempId = ""
[ExtConfig] # MidJourney和微信机器人服务 API 配置，开通此功能需要配合 chatpgt-plus-exts 项目部署
ApiURL = "" # 插件扩展 API 地址
Token = "" # 这个 token 随便填，只要确保跟 chatgpt-plus-exts 项目的 token 一样就行
[OSS] # OSS 配置，用于存储 MJ 绘画图片
Active = "local" # 默认使用本地文件存储引擎
[OSS.Local]
BasePath = "./static/upload" # 本地文件上传根路径
BaseURL = "http://localhost:5678/static/upload" # 本地上传文件根 URL 如果是线上，则直接设置为 /static/upload 即可
[OSS.Minio]
Endpoint = "" # 如 172.22.11.200:9000
AccessKey = "" # 自己去 Minio 控制台去创建一个 Access Key
AccessSecret = ""
Bucket = "chatgpt-plus" # 替换为你自己创建的 Bucket，注意要给 Bucket 设置公开的读权限，否则会出现图片无法显示。
UseSSL = false
Domain = "" # 地址必须是能够通过公网访问的，否则会出现图片无法显示。
[OSS.QiNiu] # 七牛云 OSS 配置
Zone = "z2" # 区域，z0：华东，z1: 华北，na0：北美，as0：新加坡
AccessKey = ""
AccessSecret = ""
Bucket = ""
Domain = "" # OSS Bucket 所绑定的域名，如 https://img.r9it.com
[MjConfig]
Enabled = false
UserToken = ""
BotToken = ""
GuildId = ""
ChanelId = ""
[SdConfig]
Enabled = false
ApiURL = "http://172.22.11.200:7860"
ApiKey = ""
Txt2ImgJsonPath = "res/text2img.json"
GPT-3.5
Windows
已经按照说明在config.toml里，配置了sd的相关参数
经过docker logs -f 查看，有如下报错信息
2023/10/22 23:01:52 /root/chatgpt-plus/api/handler/chatimpl/chat_handler.go:96 record not found
[1.324ms] [rows:0] SELECT * FROM chatgpt_chat_items WHERE chat_id='18e07831-8b72-4d48-8da7-892576a40e61' AND chatgpt_chat_items.deleted_at IS NULL ORDER BY chatgpt_chat_items.id LIMIT 1
2023-10-22T23:01:52.770+0800 INFO chatimpl/chat_handler.go:107 New websocket connected, IP: 104.28.213.171, Username: 13148924944
2023-10-22T23:01:54.259+0800 INFO sd/service.go:63 Consuming Task: {Id:70 SessionId:qaef98yxas06yz2r4t9jl5rhnnngdv4aifna32ztik Src:img Type:image UserId:96 Prompt:A beautiful Chinese girl riding on a tiger Params:{TaskId:task(jo8t50vpuavcc7u) Prompt:A beautiful Chinese girl riding on a tiger NegativePrompt:nsfw, paintings, cartoon, anime, sketches, low quality,easynegative,ng_deepnegative _v1 75t,(worst quality:2),(low quality:2),(normalquality:2),lowres,bad anatomy,bad hands,normal quality,((monochrome)),((grayscale)),((watermark)) Steps:20 Sampler:Euler a FaceFix:false CfgScale:7 Seed:-1 Height:1024 Width:1024 HdFix:false HdRedrawRate:0.3 HdScale:2 HdScaleAlg:ESRGAN_4x HdSteps:0} RetryCount:0}
2023-10-22T23:01:54.259+0800 ERROR sd/service.go:66 绘画任务执行失败：error with load text2img json template file: open res/text2img.json: no such file or directory
2023-10-22T23:01:57.261+0800 INFO sd/service.go:63 Consuming Task: {Id:70 SessionId:qaef98yxas06yz2r4t9jl5rhnnngdv4aifna32ztik Src:img Type:image UserId:96 Prompt:A beautiful Chinese girl riding on a tiger Params:{TaskId:task(jo8t50vpuavcc7u) Prompt:A beautiful Chinese girl riding on a tiger NegativePrompt:nsfw, paintings, cartoon, anime, sketches, low quality,easynegative,ng_deepnegative _v1 75t,(worst quality:2),(low quality:2),(normalquality:2),lowres,bad anatomy,bad hands,normal quality,((monochrome)),((grayscale)),((watermark)) Steps:20 Sampler:Euler a FaceFix:false CfgScale:7 Seed:-1 Height:1024 Width:1024 HdFix:false HdRedrawRate:0.3 HdScale:2 HdScaleAlg:ESRGAN_4x HdSteps:0} RetryCount:0}
并且这个日志信息在不停增加，也不知道怎么停止
No response
把你的配置信息粘贴到这里
GPT-3.5
Linux
使用docker-compose up -d创建chatgpt-plus-api和chatgpt-plus-web两个容器后，容器chatgpt-plus-api一直无法正常运行，用docker logs查看，提示如下
[Fx] ERROR fx.Invoke(main.main.func3()) called from:
main.main
/go/chatgpt-plus/api/main.go:85
runtime.main
/usr/local/go/src/runtime/proc.go:250
Failed: could not build arguments for function "main".main.func3
/go/chatgpt-plus/api/main.go:85:
failed to build *redis.Client:
received non-nil error from function "chatplus/store".NewRedisClient
/go/chatgpt-plus/api/store/redis.go:9:
dial tcp 127.0.0.1:6379: connect: connection refused
[Fx] ERROR Failed to start: could not build arguments for function "main".main.func3
/go/chatgpt-plus/api/main.go:85:
failed to build *redis.Client:
received non-nil error from function "chatplus/store".NewRedisClient
/go/chatgpt-plus/api/store/redis.go:9:
dial tcp 127.0.0.1:6379: connect: connection refused
2023/09/05 22:42:58 could not build arguments for function "main".main.func3 (/go/chatgpt-plus/api/main.go:85): failed to build *redis.Client: received non-nil error from function "chatplus/store".NewRedisClient (/go/chatgpt-plus/api/store/redis.go:9): dial tcp 127.0.0.1:6379: connect: connection refused
No response
把你的配置信息粘贴到这里GPT-3.5
Linux
对话时，后端报错
2023-11-22T13:45:00.615Z INFO chatimpl/openai_handler.go:33 HTTP请求完成，耗时：1.08993866s
2023-11-22T13:45:00.615Z ERROR chatimpl/chat_handler.go:150 error with decode response: invalid character '<' looking for beginning of value
2023/11/22 13:45:00 /home/www/yangjian102621/chatgpt-plus/api/handler/chatimpl/chat_handler.go:337 record not found
[1.847ms] [rows:0] SELECT * FROM chatgpt_chat_history WHERE user_id = 97.000000 AND chatgpt_chat_history.deleted_at IS NULL ORDER BY chatgpt_chat_history.id DESC LIMIT 1
前端报错：
ERROR
[object Object]
at handleError (webpack-internal:///./node_modules/_webpack-dev-server@4.15.1@webpack-dev-server/client/overlay.js:299:58)
at eval (webpack-internal:///./node_modules/_webpack-dev-server@4.15.1@webpack-dev-server/client/overlay.js:322:7)
api/chat/tokens (第二个) text 为空值时，
gpt-3.5-turbo-16k
text
返回code:1 message: "record not found"
全新安装后
Listen = "0.0.0.0:5678"
ProxyURL = ""
MysqlDns = "--------:--------@tcp(localhost:3306)/chatgpt_plus?charset=utf8mb4&parseTime=true"
StaticDir = "./static"
StaticUrl = "http://localhost/5678/static"
AesEncryptKey = "q4g17xsxkkz9ueamj01j0l2h"
WeChatBot = false
[Session]
SecretKey = "ul7ecndq0gyy2rcbwuo24tp5y394okhi7e69oxwf3zx53qxac9k9wpqehjvuzwjz"
MaxAge = 86400
[Manager]
Username = "admin"
Password = "admin123"
[Redis]
Host = "localhost"
Port = 6379
Password = ""
DB = 0
[ApiConfig]
ApiURL = ""
AppId = ""
Token = ""
[SmsConfig]
AccessKey = ""
AccessSecret = ""
Product = ""
Domain = ""
Sign = ""
CodeTempId = ""
[OSS]
Active = "local"
[OSS.Local]
BasePath = "./static/upload"
BaseURL = "http://localhost/5678/static/upload"
[OSS.Minio]
Endpoint = ""
AccessKey = ""
AccessSecret = ""
Bucket = ""
UseSSL = false
Domain = ""
[OSS.QiNiu]
Zone = ""
AccessKey = ""
AccessSecret = ""
Bucket = ""
Domain = ""
[OSS.AliYun]
Endpoint = ""
AccessKey = ""
AccessSecret = ""
Bucket = ""
Domain = ""
[MjConfig]
Enabled = false
UserToken = ""
BotToken = ""
GuildId = ""
ChanelId = ""
[SdConfig]
Enabled = false
ApiURL = ""
ApiKey = ""
Txt2ImgJsonPath = "res/text2img.json"
[XXLConfig]
Enabled = false
ServerAddr = ""
ExecutorIp = ""
ExecutorPort = ""
AccessToken = ""
RegistryKey = ""
[AlipayConfig]
Enabled = false
SandBox = false
AppId = ""
UserId = ""
PrivateKey = ""
PublicKey = ""
AlipayPublicKey = ""
RootCert = ""
NotifyURL = ""
chatgpt因为网络问题我们不够稳定，一直用的api2d,系统是否支持api2d的接口
No response
No response
# Windows对make的支持有问题, 网上提到通过chocolate来安装make的方式试了下，依然不OK。
# 最后摸索了种方案，可以在chatgpt-plus\api\go 目录下，powershell执行如下命令，可以实现打包编译的效果
# GOOS为目标环境的系统，GOARCH为目标环境的CPU架构
$env:CGO_ENABLED=0; $env:GOOS="linux" ; $env:GOARCH="amd64"; go build -o bin/chatgpt-v3-amd64-linux main.go 打包好之后，拷贝到服务器的某个目录下。第一次运行，会生成一个config.toml文件。记得修改这个文件中的配置内容为你自己的
后台运行& 关闭服务的命令
# 后台运行，避免窗口关闭，服务也跟着退出。注意改成你自己的包文件地址和要输出的日志文件地址
nohup /software/apps/chatgpt-v3-amd64-linux > /software/logs/test/chatgpt-plus.log 2>&1 &
# 基于端口号关闭进程
fuser -n tcp -k 5678GPT-3.5
Linux
配置了七牛云存储。
如果是api服务刚启动，可以正常上传头像。
过几个小时之后，再上传头像就会提示expired token！
重启服务，就又能正常几个小时。
如果选择本地存储，不存在这个问题。
支付功能是否是配置问题吗，还是目前没有这个功能
界面上没看到
服务于我微信群的粉丝，考虑运营成本，需要有一定的付费
Do you Provide an English Version, Is it possible that I Can I translate the project?
English Version
English Version
GPT-3.5
Windows
登录时提示
2023/07/29 17:48:38 /golang/src/chatgpt-plus/api/go/handler/user_handler.go:98 Error 1364 (HY000): Field 'mobile' doesn't have a default value
看了数据库有默认值
按照文档部署
把你的配置信息粘贴到这里Listen = "0.0.0.0:5678"
ProxyURL = "http://172.22.11.200:7777"
MysqlDns = "root:xx@tcp(localhost:3306)/chatgpt_plus?charset=utf8mb4&collation=utf8mb4_unicode_ci&parseTime=True&loc=Local"
StaticDir = "./static"
StaticUrl = "http://localhost:5678/static"
AesEncryptKey = "{YOUR_AES_KEY}"
StartWechatBot = false
EnabledMsgService = false
[Session]
Driver = "cookie"
SecretKey = "m0cjm3gsuw9jk73np1ni7r42koilybjcndlycjdmq7za3pbqn7w12fyok5pqh"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = "localhost"
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[Manager]
Username = "admin"
Password = "admin12345"
[Redis]
Host = "localhost"
Port = 6379
GPT-3.5
MacOS
websocket连接不上
ChatPlus.vue:307 WebSocket connection to 'wss://localhost/api/chat?sessionId=b9iluahtpcw7xduexz5e2feutk5lwpq86isjo2h0x4&role=gpt' failed:
ChatPlus.vue:307 WebSocket connection to 'wss://localhost/api/chat?sessionId=b9iluahtpcw7xduexz5e2feutk5lwpq86isjo2h0x4&role=gpt' failed:
把你的配置信息粘贴到这里
``GPT-3.5
Windows
如果 GPT 输出的内容有表情包的话，无法正确保存聊天记录。后端 API 报错：
Conversion from collation utf8mb3_general_ci into utf8mb4_0900_ai_ci impossible for parameter
No response
把你的配置信息粘贴到这里GPT-3.5
Linux
配置Azure参数后，请求失败
一般而言微软的azure需要配置的有三个值, ENDPOINT,API-KEY和DEPLOYMENT-NAME
ENDPOINT形如https://chat-bot-api.openai.azure.com, DEPLOYMENT-NAME是部署模型的自定义名称
详情参考官方链接:https://learn.microsoft.com/zh-cn/azure/ai-services/openai/quickstart?tabs=command-line&pivots=rest-api#retrieve-key-and-endpoint
在后台的系统设置里Azure api我填的是:{ENDPOINT}/openai/deployments/{DEPLOYMENT-NAME}/chat/completions?api-version=0613
（上面0613是部署的模型的版本）
API-KEY则填在后台的API-KEY管理里
模型配置：
结果：
Listen = "0.0.0.0:5678"
ProxyURL = "" # 如 http://127.0.0.1:7777
MysqlDns = "root:123456@tcp({myip})/chatgpt_plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static" # 静态资源的目录
StaticUrl = "/static" # 静态资源访问 URL
AesEncryptKey = ""
WeChatBot = false
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80" # 注意：这个是 JWT Token 授权密钥，生产环境请务必更换
MaxAge = 86400
[Manager]
Username = "admin"
Password = "admin123" # 如果是生产环境的话，这里管理员的密码记得修改
[Redis] # redis 配置信息
Host = "{myip}"
Port = 6379
Password = ""
DB = 0
[ApiConfig] # 微博热搜，今日头条等函数服务 API 配置，此为第三方插件服务，如需使用请联系作者开通
ApiURL = ""
AppId = ""
Token = ""
[SmsConfig] # 阿里云短信服务配置
AccessKey = ""
AccessSecret = ""
Product = "Dysmsapi"
Domain = "dysmsapi.aliyuncs.com"
Sign = ""
CodeTempId = ""
[ExtConfig] # MidJourney和微信机器人服务 API 配置，开通此功能需要配合 chatpgt-plus-exts 项目部署
ApiURL = "" # 插件扩展 API 地址
Token = "" # 这个 token 随便填，只要确保跟 chatgpt-plus-exts 项目的 token 一样就行
[OSS] # OSS 配置，用于存储 MJ 绘画图片
Active = "local" # 默认使用本地文件存储引擎
[OSS.Local]
BasePath = "./static/upload" # 本地文件上传根路径
BaseURL = "http://localhost:5678/static/upload" # 本地上传文件根 URL 如果是线上，则直接设置为 /static/upload 即可
[OSS.Minio]
Endpoint = "" # 如 172.22.11.200:9000
AccessKey = "" # 自己去 Minio 控制台去创建一个 Access Key
AccessSecret = ""
Bucket = "chatgpt-plus" # 替换为你自己创建的 Bucket，注意要给 Bucket 设置公开的读权限，否则会出现图片无法显示。
UseSSL = false
Domain = "" # 地址必须是能够通过公网访问的，否则会出现图片无法显示。
[OSS.QiNiu] # 七牛云 OSS 配置
Zone = "z2" # 区域，z0：华东，z1: 华北，na0：北美，as0：新加坡
AccessKey = ""
AccessSecret = ""
Bucket = ""
Domain = "" # OSS Bucket 所绑定的域名，如 https://img.r9it.com
[MjConfig]
Enabled = false
UserToken = ""
BotToken = ""
GuildId = ""
ChanelId = ""
[SdConfig]
Enabled = false
ApiURL = "{sdip}"
ApiKey = ""
Txt2ImgJsonPath = "res/text2img.json"
GPT-3.5
Linux
按照安装教程部署在海外节点（韩国）后，页面能正常显示，能获取到角色及收到"本项目已开放全部源代码：https://github.com/yangjian102621/chatgpt-plus..."等信息，但是发送消息却无法收到返回信息。
1，排查了Key没问题；
2，对比了Toml配置文件没看出异常。
运行日志：
2023-05-12T05:09:05.440Z INFO src/main.go:20 Loading config file: config.toml
2023-05-12T05:09:05.442Z INFO server/server.go:152 http://0.0.0.0:5678
2023-05-12T05:09:14.748Z INFO server/handler_chat.go:40 New websocket connected, IP: 127.0.0.1:39862, Username: Jran
2023-05-12T05:09:19.170Z INFO server/handler_chat.go:67 Receive a message: 1111
2023-05-12T05:09:19.170Z INFO server/handler_chat.go:132 会话上下文：[]
2023-05-12T05:09:19.170Z INFO server/handler_chat.go:179 Use API KEY: sk-Kxxx, PROXY:
2023-05-12T05:09:19.369Z ERROR server/handler_chat.go:213 EOF
chatplus/server.(*Server).sendMessage
/Users/ran/Developer/codes/chatgpt-plus/src/server/handler_chat.go:213
chatplus/server.(*Server).ChatHandle.func2
/Users/ran/Developer/codes/chatgpt-plus/src/server/handler_chat.go:73
2023-05-12T05:09:19.369Z INFO server/handler_chat.go:78 回答完毕: 1111
Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "0.0.0.0:5678"
ProxyURL = [""]
AccessKey = "test"
[Session]
SecretKey = "test001"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = "chat.xxx.com"
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = ""
WechatGroup = ""
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 1.0
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
[[Chat.ApiKeys]]
Value = "sk-xxxxxxx"
LastUsed = 1683867437我看了下平台readme有介绍说支持文心一言，但是我在添加key的时候没找到文心一言的添加方式。想问下平台支持文心一言的接入吗？
No response
No response
如图
No response
No response
GPT-3.5
Windows
使用arm64架构对api进行打包编译，制作镜像运行容器后，访问主页为空白页，访问容器日志无提示报错
No response
把你的配置信息粘贴到这里GPT-3.5
Linux
环境：
Vultr大阪，debian-11.7，内核5.10.0-25-amd64，MySQL 8.2.0，Redis 6.2.14
部署版本：3.1.7
按部署文档逐步部署完成后，chat-gpt-plus-api日志一直提示如下错误，并重复重启：
Table 'gpt-plus.chatgpt_configs' doesn't exist
按照文档逐步部署，并启动docker拉取镜像后发生。
Listen = "0.0.0.0:5678"
ProxyURL = "" # 如 http://127.0.0.1:7777
MysqlDns = "root:------------@tcp(mysql:3306)/gpt-plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static" # 静态资源的目录
StaticUrl = "/static" # 静态资源访问 URL
AesEncryptKey = ""
WeChatBot = false
[Session]
SecretKey = "------------------------------------g80" # 注意：这个是 JWT Token 授权密钥，生产环境请务必更换
MaxAge = 86400
[Manager]
Username = "admin"
Password = "admin123" # 如果是生产环境的话，这里管理员的密码记得修改
[Redis] # redis 配置信息
Host = "redis"
Port = 6379
Password = "------------------GP7"
DB = 0
sd如果输入中文，生成效果很差，是否可以支持将prompt翻译后再生成呢
加个翻译接口
加个翻译接口
加个翻译接口
GPT-3.5
Linux
查看了docker日志，部署成功的，也没有报错，数据库也可以链接上去，就是登录时候一直报错502
No response
把你的配置信息粘贴到这里GPT-3.5
Windows
main.go没有任何更改
go run main.go 运行报错：main.go:14:12: pattern dist: no matching files found
No response
把你的配置信息粘贴到这里Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "0.0.0.0:5678"
ProxyURL = []
AccessKey = "LTAI5tXXXXXXXX"
[Session]
SecretKey = "CGsTfjMgeXXXXXXXX"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = "172.XXXXXX"
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = ""
WechatGroup = ""
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 1.0
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
[[Chat.ApiKeys]]
Value = "sk-IOdebXXXXXXXXXXXXXXXXXXXXXXX"
LastUsed = 0
GPT-3.5
Windows
打开windows的快速建构包进入 localhost:5678 就是这样
一直都是这样
自动生成的配置
GPT-4
MacOS
有时候出现，有时候不出现。
把你的配置信息粘贴到这里GPT-3.5
Linux
大佬你好，我是在虚拟机上用docker部署的，安装部署过程中没有报错，登录时报错了,前端移动端我都试过了，报同样的错误
我是在虚拟机上用docker部署的，安装运行过程中都没有报错，登录后台时报错了
Listen = "0.0.0.0:5678"
#ProxyURL = "http://172.22.11.200:7777"
ProxyURL = "http://127.0.0.1:7777"
MysqlDns = "root:12345678@tcp(172.28.173.76:3307)/chatgpt_plus?charset=utf8mb4&parseTime=True&loc=Local"
StaticDir = "./static"
StaticUrl = "http://localhost:8080/static"
[Session]
Driver = "cookie"
SecretKey = "8k0c67y2or1n7kbmn1w1c86ygqscguoktuf9t524jm64ls585z8uibpdssiy128s"
Name = "CHAT_PLUS_SESSION"
Path = "/"
Domain = ""
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[Manager]
Username = "admin"
Password = "admin_1234"
[Redis]
Host = "localhost"
Port = 6379
Password = ""
系统版本是centos 7，内核是3.10.0-1062.el7.x86_64，内存为8G，cpu为2核，硬盘容量128G
GPT-3.5
Windows
前端登录后(http://127.0.0.1:8888/chat)，无法连接后端websocket。判断应该是后端websocket不生效。
1.登录http://127.0.0.1:8888/chat
2.加载后，右边遮罩无法关闭。控制台websocket连接报错。
把你的配置信息粘贴到这里GPT-3.5
Windows
编译部署完成后无法连接后端，显示{"code":400,"message":"Not Authorized"}
多次尝试后还是无法登录
No response
server {
listen 8080;
server_name www.chatgpt.com; #替换成你自己的域名
#ssl_certificate xxx.pem; # 替换成自己的 SSL 证书
#ssl_certificate_key xxx.key;
#ssl_session_timeout 5m;
#ssl_ciphers ECDHE-RSA-AES128-GCM-SHA256:ECDHE:ECDH:AES:HIGH:!NULL:!aNULL:!MD5:!ADH:!RC4;
#ssl_protocols TLSv1 TLSv1.1 TLSv1.2;
#ssl_prefer_server_ciphers on;
# 日志地址
access_log /var/log/chatgpt/access.log;
error_log /var/log/chatgpt/error.log;
index index.html;
root /var/www/chatgpt/dist; # 这里改成前端静态页面的地址
location / {
try_files $uri $uri/ /index.html;
# 后端 API 的转发
location /api/ {
proxy_http_version 1.1;
proxy_connect_timeout 300s;
proxy_read_timeout 300s;
proxy_send_timeout 12s;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_pass http://ip:5678; # 这里改成后端服务的内网 IP 地址
}
# 静态资源转发
location /static/ {
proxy_pass http://ip:5678; # 这里改成后端服务的内网 IP 地址
}
}
}
GPT-4
Linux
云服务器
192.168.0.129为司有IP
前后端分类部署，
后端、前端服务都启动
可重现11111111111
Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "192.168.0.129:5678"
ProxyURL = ["YOUR_PROXY_URL"]
AccessKey = "YOUR_ACCESS_KEY"
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = "192.168.0.129"
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = "https://img.r9it.com/chatgpt/WX20230505-162403.png"
WechatGroup = " https://img.r9it.com/chatgpt/WX20230505-162538.png"
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 1.0
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
[[Chat.ApiKeys]]
Value = "1233333333333333333333333333333333333333333333333333"
LastUsed = 0
VUE_APP_API_HOST=http://192.168.0.129:5678
VUE_APP_WS_HOST=ws://192.168.0.129:5678
GPT-3.5
Linux
registry.cn-hangzhou.aliyuncs.com/
docker-compose up -d
Pulling chatgpt-plus-api (registry.cn-hangzhou.aliyuncs.com/geekmaster/chatgpt-plus-api:v3.0.5.2)...
ERROR: The image for the service you're trying to recreate has been removed. If you continue, volume data could be lost. Consider backing up your data before continuing.
把你的配置信息粘贴到这里docker-compose up -d
Pulling chatgpt-plus-api (registry.cn-hangzhou.aliyuncs.com/geekmaster/chatgpt-plus-api:v3.0.5.2)...
ERROR: The image for the service you're trying to recreate has been removed. If you continue, volume data could be lost. Consider backing up your data before continuing.
允许用户修改登录密码，采用弹框的交互形式
No response
No response
GPT-3.5
Windows
后台可以登录，前台输入口令后一直转圈圈
code包，运行后就这样
Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "192.168.11.138:5678"
ProxyURL = ["127.0.0.1:7890"]
AccessKey = "YOUR_ACCESS_KEY"
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = "192.168.11.138"
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = "https://img.r9it.com/chatgpt/WX20230505-162403.png"
WechatGroup = " https://img.r9it.com/chatgpt/WX20230505-162538.png"
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 1.0
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
GPT-3.5
Windows
SD绘图每次提交一张额度就会减一，看上去没啥问题。实际刷新浏览器额度又回来了。
sd刷新浏览器或重新登录额度又回来了。
把你的配置信息粘贴到这里GPT-3.5
Windows
2023-10-10T17:34:17.304+0800 INFO mj/service.go:61 Consuming Task: {Id:0 SessionId:mixjipz7cs7lyizs70q1wtumcp3uqy1vu9qt5dfh4i Src:chat Type:image UserId:4 Prompt:Elon Musk driving a tractor in the countryside of China in the 20th century. ChatId:61d5c5a1-1346-40dd-8ecf-af915ba4e0cd RoleId:1 Icon:/images/avatar/mid_journey.png Index:0 MessageId: MessageHash: RetryCount:0}
2023-10-10T17:34:17.604+0800 ERROR mj/service.go:74 绘画任务执行失败：error with http request: %!w()
2023-10-10T17:34:20.606+0800 INFO mj/service.go:61 Consuming Task: {Id:0 SessionId:mixjipz7cs7lyizs70q1wtumcp3uqy1vu9qt5dfh4i Src:chat Type:image UserId:4 Prompt:Elon Musk driving a tractor in the countryside of China in the 20th century. ChatId:61d5c5a1-1346-40dd-8ecf-af915ba4e0cd RoleId:1 Icon:/images/avatar/mid_journey.png Index:0 MessageId: MessageHash: RetryCount:0}
2023-10-10T17:34:20.861+0800 ERROR mj/service.go:74 绘画任务执行失败：error with http request: %!w()
我没配置mj，或者在mj生成的时候点击取消。然后系统后台日志一直报这个。会尝试多少次请求然后取消？
把你的配置信息粘贴到这里GPT-3.5
Windows
如题 纯小白一只，烦请大佬解答！、
按照b站视频教程部署，就是进页面就闪退回主页面
location /api/ {
proxy_http_version 1.1;
proxy_connect_timeout 300s;
proxy_read_timeout 300s;
proxy_send_timeout 12s;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_pass http://101.37.167.132:5678; # 这里改成后端服务的内网 IP 地址
}
# 静态资源转发
location /static/ {
proxy_pass http://101.37.167.132:5678; # 这里改成后端服务的内网 IP 地址
}
}
}
Listen = "0.0.0.0:5678"
ProxyURL = "http://127.0.0.1:7777"
MysqlDns = "root:12345678@tcp(101.37.167.132:3307)/chatgpt_plus?charset=utf8mb4&collation=utf8mb4_unicode_ci&parseTime=True&loc=Local"
StaticDir = "./static"
StaticUrl = "/static"
AesEncryptKey = "{YOUR_AES_KEY}"
StartWechatBot = false
EnabledMsgService = false
GPT-3.5
Linux
输入账号密码后，登录没有反应，浏览器提示
No response
把你的配置信息粘贴到这里GPT-3.5
Linux
找不到 xx.xx.x.x的网页找不到与以下网址对应的网页：http://x.x.x.x:5678/chat/
HTTP ERROR 404
我的8888端口被占用了，前端的端口我改成了8889
几乎是按照文档默认的配置安装的
Title = "Chat-Plus AI 助手"
ConsoleTitle = "Chat-Plus 控制台"
Listen = "0.0.0.0:5678"
ProxyURL = [""]
AccessKey = ""
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80"
Name = "CHAT_SESSION_ID"
Path = "/"
Domain = ""
MaxAge = 86400
Secure = false
HttpOnly = false
SameSite = 2
[ImgURL]
WechatCard = "https://img.r9it.com/chatgpt/WX20230505-162403.png"
WechatGroup = " https://img.r9it.com/chatgpt/WX20230505-162538.png"
[Manager]
Username = "admin"
Password = "admin123"
[Chat]
ApiURL = "https://api.openai.com/v1/chat/completions"
Model = "gpt-3.5-turbo"
Temperature = 1.0
MaxTokens = 1024
EnableContext = true
ChatContextExpireTime = 3600
[[Chat.ApiKeys]]
Value = "sk-xx"
LastUsed = 0
服务正常启动
2023-05-20T04:42:19.134+0800 INFO src/main.go:18 Loading config file: config.toml
2023-05-20T04:42:19.135+0800 INFO server/server.go:136 Loading config file:aa dist
2023-05-20T04:42:19.135+0800 INFO server/server.go:153 http://0.0.0.0:5678将聊天页面的功能拆分成组件，然后每个组件分别做移动端 UI 实现，聊天的交互核心逻辑共用，这样可以做最小改动。
No response
不用额外在开发移动版了
GPT-3.5
MacOS
配置好 mj后绘画完成历史记录里的图片无法正常显示
No response
Listen = "0.0.0.0:5678"
ProxyURL = "" # 如 http://127.0.0.1:7777
MysqlDns = "root:12345678@tcp(10.0.0.227:3307)/chatgpt_plus?charset=utf8&parseTime=True&loc=Local"
StaticDir = "./static" # 静态资源的目录
StaticUrl = "/static" # 静态资源访问 URL
AesEncryptKey = ""
WeChatBot = false
[Session]
SecretKey = "azyehq3ivunjhbntz78isj00i4hz2mt9xtddysfucxakadq4qbfrt0b7q3lnvg80" # 注意：这个是 JWT Token 授权密钥，生产环境请务必更换
MaxAge = 86400
[Manager]
Username = "admin"
Password = "admin123" # 如果是生产环境的话，这里管理员的密码记得修改
[Redis] # redis 配置信息
Host = "10.0.0.227"
Port = 6379
Password = ""
DB = 0
[ApiConfig] # 微博热搜，今日头条等函数服务 API 配置，此为第三方插件服务，如需使用请联系作者开通
ApiURL = ""
AppId = ""
Token = ""
[SmsConfig] # 阿里云短信服务配置
AccessKey = ""
AccessSecret = ""
Product = "Dysmsapi"
Domain = "dysmsapi.aliyuncs.com"
Sign = ""
CodeTempId = ""
[ExtConfig] # MidJourney和微信机器人服务 API 配置，开通此功能需要配合 chatpgt-plus-exts 项目部署
ApiURL = "" # 插件扩展 API 地址
Token = "" # 这个 token 随便填，只要确保跟 chatgpt-plus-exts 项目的 token 一样就行
[OSS] # OSS 配置，用于存储 MJ 绘画图片
Active = "minio" # 默认使用本地文件存储引擎
[OSS.Local]
BasePath = "./static/upload" # 本地文件上传根路径
BaseURL = "http://10.0.0.227:5678/static/upload" # 本地上传文件根 URL 如果是线上，则直接设置为 /static/upload 即可
[OSS.Minio]
Endpoint = "http://mydomain:9000" # 如 172.22.11.200:9000
AccessKey = "YG8xX0bjttgnqktgBY" # 自己去 Minio 控制台去创建一个 Access Key
AccessSecret = "1twhZ0kd9IFl91N5eJUTyMNyGMEO0VasG"
Bucket = "chatgpt-plus" # 替换为你自己创建的 Bucket，注意要给 Bucket 设置公开的读权限，否则会出现图片无法显示。
UseSSL = false
Domain = "http://mydomain:9000" # 地址必须是能够通过公网访问的，否则会出现图片无法显示。
[OSS.QiNiu] # 七牛云 OSS 配置
Zone = "z2" # 区域，z0：华东，z1: 华北，na0：北美，as0：新加坡
AccessKey = ""
AccessSecret = ""
Bucket = ""
Domain = "" # OSS Bucket 所绑定的域名，如 https://img.r9it.com
[MjConfig]
Enabled = true
UserToken = "MTE1NDYyNzM0ODAzODE3NjgwOA.GICqlF.N2zX1i621Qsz"#已经删除部分
BotToken = "MTE1NTY5ODgwNzE4MSwnuruBK-5AsZqXjhPet0xRE"#已经删除部分
GuildId = "1155691"#已经删除部分
ChanelId = "115569"#已经删除部分
[SdConfig]
Enabled = false
ApiURL = "http://172.22.11.200:7860"
ApiKey = ""
Txt2ImgJsonPath = "res/text2img.json"GPT-3.5
Windows
后台新增【角色】后，前台的下拉菜单中不显示
编辑角色
角色名称：正则表达式
角色标志：regex
角色图标：/images/avatar/elon_musk.jpg
打招呼信息：我是正则表达式生成师
上下文信息：
user 我希望你充当正则表达式生成器。您的角色是生成匹配文本中特定模式的正则表达式。您应该以一种可以轻松复制并粘贴到支持正则表达式的文本编辑器或编程语言中的格式提供正则表达式。不要写正则表达式如何工作的解释或例子；只需提供正则表达式本身。
启用状态：On
排序：18
把你的配置信息粘贴到这里
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.