chrisyy2003 / nonebot-plugin-gpt3 Goto Github PK

大佬你好，想问一下能不能关闭预置人格？

就在今天，OPENai更新了gpt-3.5-turbo语言模型

我这里全部都配置好之后请求聊天，但是chatgpt接口返回提示：

 发生错误: Incorrect API key provided: sk-hZr8S***************************************iYP8. You can find your API key at https://beta.openai.com.

我确认我的key完全正确没有错误，所以这是什么鬼，重新生成了好几次也还是key返回错误
你们有遇见这种怪问题吗
怎么解决的

02-25 16:12:14 [ERROR] nonebot | Failed to import "nonebot_plugin_gpt3"
Traceback (most recent call last):
File "", line 14, in
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/load.py", line 122, in load_from_toml
return load_all_plugins(plugins, plugin_dirs)
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/load.py", line 59, in load_all_plugins
return manager.load_all_plugins()
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/manager.py", line 168, in load_all_plugins
return set(
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/manager.py", line 169, in
filter(None, (self.load_plugin(name) for name in self.available_plugins))

File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/manager.py", line 141, in load_plugin
module = importlib.import_module(name)
File "/usr/lib/python3.9/importlib/init.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1030, in _gcd_import
File "", line 1007, in _find_and_load
File "", line 986, in _find_and_load_unlocked
File "", line 680, in _load_unlocked
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot/plugin/manager.py", line 233, in exec_module
super().exec_module(module)
File "", line 790, in exec_module
File "", line 228, in _call_with_frames_removed
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot_plugin_gpt3/init.py", line 10, in
from .config import *
File "/opt/openai-bot/.venv/lib/python3.9/site-packages/nonebot_plugin_gpt3/config.py", line 24, in
config = Config.parse_obj(global_config)
File "pydantic/main.py", line 527, in pydantic.main.BaseModel.parse_obj
return cls(**obj)
File "pydantic/env_settings.py", line 39, in pydantic.env_settings.BaseSettings.init
super().init(
File "pydantic/main.py", line 342, in pydantic.main.BaseModel.init
raise validation_error
pydantic.error_wrappers.ValidationError: 1 validation error for Config
gpt3_api_key_list_from_env
value is not a valid list (type=type_error.list)

系统环境&软件版本：Python3.9、Debian11、NoneBot2

rt

你没 require 快加上！不然会爆炸的！

今天刚搭好就一直在报错，不知道该怎么办😭

用pip安装插件时下载完成但是无法安装
pip install 这个插件时下载的了，最后没有安装
其次我在隔壁的大佬那里白嫖到了免费API的方法，奈何自己毫无语言基础
求大佬能增加一个免费API的功能
下面是大佬写出的东西
免KEY接口分析.docx

坏了，昨天开始，bot的回复成这样了

但是chatgpt网站还能用 能对接token吗

一升rc版本插件都炸完了:(

按照主页的截图测试的，不知道是不是只有我存在这个问题

现在国内ip似乎 api也不给用了

03-08 16:43:12 [SUCCESS] nonebot | ONEBOT V11 2545044071 | [message.group.normal]: Message 470695935 from 1270217000@[群:925353737] 'gpt 原神是否有抄袭塞尔达'
03-08 16:43:12 [INFO] nonebot | Event message.group.normal is ignored

• 插件版本：1.1.3
• openai版本：0.27.0
• 模型版本：gpt-3.5-turbo

在聊天开始时，会有概率出现请求token长度大于模型最大token长度的情况，如下图所示。

是密钥有问题吗

建议新增个人维度的每天提问次数CD限制，方便控制频率，不然免费API额度很快会被用完

powershell启动go-cqhttp的时候提示账号密码错误或账号被冻结，但是我账号密码都正确，而且扫码登录也是相同的提示

向机器人发送设置人格的指令后机器人不处理指令。而重置人格的指令可以正常处理

服务器过载后会话报错：
发生错误: The server is overloaded or not ready yet.
或者：

之后对话就重置了，请问有解决方法吗，比如多设置几个key

设置人格、重置会话等操作需要为管理员

发送消息，机器人回复error，是哪里出错了。日志如下

06-27 15:22:48 [SUCCESS] nonebot | OneBot V11 3568483338 | [message.private.friend]: Message 971437277 from '.xxx'
06-27 15:22:48 [DEBUG] nonebot | Checking for matchers in priority 10...
06-27 15:22:48 [DEBUG] nonebot | Checking for matchers in priority 100...
06-27 15:22:48 [INFO] nonebot | Event will be handled by Matcher(type='message', module=nonebot_plugin_gpt3)
06-27 15:22:48 [DEBUG] nonebot | Running Matcher(type='message', module=nonebot_plugin_gpt3)
06-27 15:22:48 [DEBUG] nonebot | Running handler Dependent(call=_)
06-27 15:22:49 [DEBUG] nonebot | OneBot V11 | Calling API send_msg
06-27 15:22:49 [INFO] nonebot | Matcher(type='message', module=nonebot_plugin_gpt3) running complete
06-27 15:22:49 [DEBUG] nonebot | Stop event propagation
06-27 15:22:49 [DEBUG] nonebot | Checking for matchers completed

为什么我的gpt3说今年是2020年，是OpenAI开放的是2020年的库吗

关于人格设置 我输入了人格设置 猫娘 机器人给的反应是全局人格设置成功 但是其实人格还是AI助手 而且有时候设置还会出现双重人格
我是AI猫娘这种情况... 不知道是我的问题还是此插件的bug 我已经重新卸载安装后依然如此 而且不知道该插件是否支持自定义人格保存呢

02-22 18:49:15 [INFO] nonebot_plugin_gocqhttp | [机器人QQ] 收到群 群名(群号) 内 群友(群友QQ号) 的消息: /gpt3 (-562778938)
02-22 18:49:15 [SUCCESS] nonebot | OneBot V11 3463402239 | [message.group.normal]: Message -562778938 from 群友QQ号@[群:群号] '/gpt3'
02-22 18:49:15 [INFO] nonebot | Event will be handled by Matcher(type='message', module=nonebot_plugin_gpt3)
02-22 18:49:15 [INFO] nonebot | Matcher(type='message', module=nonebot_plugin_gpt3) running complete

使用pip install nonebot-plugin-gpt3 playwright -U安装插件并在bot.py里加上nonebot.load_plugin('nonebot_plugin_gpt3')后，bot无法启动，发生如下报错

已经下载了README.md里的rustup-steup.exe并安装，但是没有手动配置环境变量，在cmd中输入rustup -V后有以下输出：
rustup 1.25.1 (bb60b1e89 2022-07-12)
info: This is the version for the rustup toolchain manager, not the rustc compiler.
info: The currently active rustc version is rustc 1.66.0 (69f9c33d7 2022-12-12)

发生错误: Request timed out: HTTPSConnectionPool(host='api.openai.com', port=443): Max retries exceeded with url: /v1/chat/completions (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x000002406F0B8C10>, 'Connection to api.openai.com timed out. (connect timeout=600)'))

总是回答：作为AI语言模型，没有真正的身份和个性

使用插件时出现以下报错：
“You requested a model that is not compatible with this engine.”
与https://community.openai.com/t/error-you-requested-a-model-that-is-not-compatible-with-this-engine-when-using-codex-in-playground/35153/1 存在相同的问题。

自己配置好所有的文件之后跑起来报错未配置apikey，检查之后apikey是已经填写并保存的状态，劳烦大佬看看是什么问题
以下是运行时控制台的显示内容
(kugel) D:\kugelblyat\mai-bot-main>python bot.py
04-13 08:45:54 [SUCCESS] nonebot | NoneBot is initializing...
04-13 08:45:54 [INFO] nonebot | Current Env: dev
04-13 08:45:55 [WARNING] nonebot_plugin_gpt3 | 没有配置api key
04-13 08:45:55 [INFO] nonebot_plugin_gpt3 | 加载代理: 无
04-13 08:45:55 [INFO] nonebot_plugin_gpt3 | 加载默认人格:
04-13 08:45:55 [SUCCESS] nonebot | Succeeded to load plugin "nonebot_plugin_gpt3"
04-13 08:45:57 [SUCCESS] nonebot | Succeeded to load plugin "maimaidx" from "src.plugins.maimaidx"
04-13 08:45:57 [SUCCESS] nonebot | Succeeded to load plugin "public" from "src.plugins.public"
04-13 08:45:57 [SUCCESS] nonebot | Running NoneBot...
04-13 08:45:57 [INFO] uvicorn | Started server process [2588]
04-13 08:45:57 [INFO] uvicorn | Waiting for application startup.
04-13 08:45:57 [INFO] uvicorn | Application startup complete.
04-13 08:45:57 [INFO] uvicorn | Uvicorn running on http://127.0.0.1:10219 (Press CTRL+C to quit)
04-13 08:46:00 [INFO] uvicorn | ('127.0.0.1', 60610) - "WebSocket /onebot/v11/ws" [accepted]
04-13 08:46:02 [INFO] nonebot | OneBot V11 | Bot 1976405702 connected
04-13 08:46:02 [INFO] websockets | connection open
apikey文件位置正确里面的api也是自己填写好了的

逆向库地址：https://github.com/acheong08/ChatGPT
如果可以，能不能加入accessToken，SessionToken认证登录呢？

回复后触发，后台无报错

我的nonebot是部署在docker中的，不知道是否可以从环境变量中读取配置

01-11 12:56:36 [SUCCESS] nonebot | OneBot V11 738386129 | [message.private.friend]: Message 1516341372 from 3258447421 "/gpt3 你好"
01-11 12:56:36 [INFO] nonebot | Event will be handled by Matcher(type='message', module=nonebot_plugin_gpt3)
01-11 12:56:36 [INFO] nonebot | Matcher(type='message', module=nonebot_plugin_gpt3) running complete
01-11 12:56:36 [ERROR] nonebot | Running Matcher(type='message', module=nonebot_plugin_gpt3) failed.
Traceback (most recent call last):
File "c:\Zsy_files\Zsy_bot\ghcp\Zsybot\bot.py", line 59, in
nonebot.run(app="mp_main:app")
File "C:\Zsy_files\Python\lib\site-packages\nonebot_init_.py", line 273, in run
get_driver().run(*args, **kwargs)
File "C:\Zsy_files\Python\lib\site-packages\nonebot\drivers\fastapi.py", line 172,
in run
uvicorn.run(
File "C:\Zsy_files\Python\lib\site-packages\uvicorn\main.py", line 569, in run
server.run()
File "C:\Zsy_files\Python\lib\site-packages\uvicorn\server.py", line 60, in run
return asyncio.run(self.serve(sockets=sockets))
File "C:\Zsy_files\Python\lib\asyncio\runners.py", line 44, in run
return loop.run_until_complete(main)
File "C:\Zsy_files\Python\lib\asyncio\base_events.py", line 636, in run_until_complete
self.run_forever()
File "C:\Zsy_files\Python\lib\asyncio\windows_events.py", line 321, in run_forever
super().run_forever()
File "C:\Zsy_files\Python\lib\asyncio\base_events.py", line 603, in run_forever
self._run_once()
File "C:\Zsy_files\Python\lib\asyncio\base_events.py", line 1899, in _run_once
handle._run()
File "C:\Zsy_files\Python\lib\asyncio\events.py", line 80, in _run
self._context.run(self._callback, *self._args)
File "C:\Zsy_files\Python\lib\site-packages\nonebot\message.py", line 142, in _check_matcher
await _run_matcher(Matcher, bot, event, state, stack, dependency_cache)

File "C:\Zsy_files\Python\lib\site-packages\nonebot\message.py", line 188, in run_matcher
await matcher.run(bot, event, state, stack, dependency_cache)
File "C:\Zsy_files\Python\lib\site-packages\nonebot\internal\matcher\matcher.py", line 727, in run
await self.simple_run(bot, event, state, stack, dependency_cache)
File "C:\Zsy_files\Python\lib\site-packages\nonebot\internal\matcher\matcher.py", line 702, in simple_run
await handler(
File "C:\Zsy_files\Python\lib\site-packages\nonebot\dependencies_init.py", line 108, in call
return await cast(Callable[..., Awaitable[R]], self.call)(**values)
File "C:\Zsy_files\Python\lib\site-packages\nonebot_plugin_gpt3_init_.py", line
169, in _
resp = await get_user_session(session_id).get_chat_response(msg)
File "C:\Zsy_files\Python\lib\site-packages\nonebot_plugin_gpt3_init_.py", line
81, in get_chat_response
if ok:
UnboundLocalError: local variable 'ok' referenced before assignment

配置文件：
GPT3_API_KEY_LIST_FROM_ENV=["sk-****rMrwn"]
gpt3_model = 'gpt-3.5-turbo' # 语言模型
gpt3_command_prefix = "c" # 基本触发命令
gpt3_need_at = False # 是否需要@才触发命令
gpt3_image_render = True # 是否需要将回答渲染为图片
gpt3_image_limit = 100 # 长度超过多少才会渲染成图片
gpt3_max_tokens = 400 # 返回回答的最大长度

这样之后就没动静了

是网络问题吗？

插件已经加载了，key已经输入了，切换人格的什么都能用，但是那个对话消息就是触发不了，后台好像也显示没有反应

发生错误: Error communicating with OpenAI: HTTPSConnectionPool(host='api.openai.com', port=443): Max retries exceeded with url: /v1/chat/completions (Caused by NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x0000029AA7BEEB20>: Failed to establish a new connection: [WinError 10060] 由于连接方在一段时间后没有正确答复或连接的主机没有反应，连接尝试失败。'))

大佬你好，请问一下这个报错是因为API被墙的缘故吗，配置文件里可以支持写代理吗

发生未知错误：<calss 'anyio.EndOfStream'>

配置文件添加一个时间参数，超过x秒后自动结束聊天并提醒会话超时聊天结束（？）

text-chat-davinci-002-20230126