waylaidwanderer / node-chatgpt-api Goto Github PK

Hi,bro can you tell me how to solve 405?thanks a lot

please add "user" parameter

It seems there is no timeout option like timeoutMs for the function. Sometimes a request gets stuck for a long time。

hi, how to use it with commonjs?

thx

It looks like the model is no longer available...

I follow the get start guide and use reverse proxy already. Do i miss any step in order to get this work ??

In the debug log

Request :
https://chatgpt.pawan.krd/api/completions

{
  model: 'text-chat-davinci-002-sh-alpha-aoruigiofdj83',
  temperature: 0.8,
  top_p: 1,
  presence_penalty: 1,
  stop: [ '<|im_end|>', '<|im_sep|>', '\nUser:' ],
  max_tokens: 1000,
  prompt: '\n' +
    '<|im_sep|>Instructions:\n' +
    'You are ChatGPT, a large language model trained by OpenAI.\n' +
    'Current date: February 13, 2023<|im_sep|>\n' +
    '\n' +
    'User:\n' +
    'Hello!<|im_end|>\n' +
    'ChatGPT:\n'
}

Response :

{"detail":"An error ocurred. Try refreshing your browser."}

using reverse proxy, some questions throw an error

TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:14152:11)
    at async ChatGPTClient.getCompletion (file:///C:/Users/JerryWn/AppData/Local/Programs/node-v18.14.0-win-x64/node_modules/@waylaidwanderer/chatgpt-api/src/ChatGPTClient.js:138:26)
    at async ChatGPTClient.sendMessage (file:///C:/Users/JerryWn/AppData/Local/Programs/node-v18.14.0-win-x64/node_modules/@waylaidwanderer/chatgpt-api/src/ChatGPTClient.js:196:28)
    at async Object.<anonymous> (file:///C:/Users/JerryWn/AppData/Local/Programs/node-v18.14.0-win-x64/node_modules/@waylaidwanderer/chatgpt-api/bin/server.js:100:18) {
  cause: SocketError: other side closed
      at TLSSocket.onSocketEnd (node:internal/deps/undici/undici:9227:26)
      at TLSSocket.emit (node:events:525:35)
      at endReadableNT (node:internal/streams/readable:1359:12)
      at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
    code: 'UND_ERR_SOCKET',
    socket: {
      localAddress: '198.18.0.1',
      localPort: 58538,
      remoteAddress: '198.18.0.10',
      remotePort: 443,
      remoteFamily: 'IPv4',
      timeout: undefined,
      bytesWritten: 1764,
      bytesRead: 0
    }
  }
}

ReferenceError: fetch is not defined
at file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/node_modules/@fortaine/fetch-event-source/lib/esm/fetch.js:46:85
at new Promise ()
at fetchEventSource (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/node_modules/@fortaine/fetch-event-source/lib/esm/fetch.js:18:12)
at file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/src/ChatGPTClient.js:77:27
at new Promise ()
at ChatGPTClient.getCompletion (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/src/ChatGPTClient.js:74:20)
at ChatGPTClient.sendMessage (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/src/ChatGPTClient.js:160:24)
at processTicksAndRejections (node:internal/process/task_queues:96:5)
at async onMessage (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/bin/cli.js:142:26)
at async file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/bin/cli.js:83:1

I use your library when I call to get the answer and only get null text results

{
author: 'bot',
id: '4e9561bb-7e67-471f-b950-dc028e8729ed',
conversationId: undefined,
clientId: undefined,
conversationSignature: undefined,
invocationId: '2',
text: ''
}

There are lots of breaking changes with the API necessitating fast paced releases upstream. Some sort of automation is required to keep pace with them and keep this package usable.

Dependabot only updates once a day so is quite slow, there must be a better method, do you know of one?

没法用了哦

Question again!

I am still on waitlist. My user id on cookie is not working....
Does this work only for people who can start using Bing Chat?
Is there any way to get this around?

Seems like we are being played by OpenAI 😆

Thats a great news that you have found out the way to use official models.
Could we possibly get a the implementation for the proxy? As using someone else free proxy from github will just bring capacity sooner or later. On another side having an option for where and how your data proccessed is great.

The reverse proxy servers worked great, but as of just now both of the two reverse proxy servers are down now, which seems inevitable given how many requests they would handle.
This happened when using cli, 2 hours ago it was working fine.

TypeError: fetch failed
    at Object.fetch (node:internal/deps/undici/undici:14152:11)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async create (file:///root/node-chatgpt-api/node_modules/@waylaidwanderer/fetch-event-source/lib/esm/fetch.js:52:34) {
  cause: Error: connect EHOSTUNREACH 2a09:8280:1::3:e03e:443
      at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1494:16) {
    errno: -113,
    code: 'EHOSTUNREACH',
    syscall: 'connect',
    address: '2a09:8280:1::3:e03e',
    port: 443
  }

`newuser@work % npm run cli

@waylaidwanderer/[email protected] cli
node bin/cli.js

╔═══════════════╗
║ ChatGPT CLI ║
╚═══════════════╝

Type "!" to access the command menu.
? Write a message: hi
TypeError: stream.getReader is not a function
at getBytes (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/node_modules/@fortaine/fetch-event-source/lib/esm/parse.js:2:27)
at create (file:///Users/newuser/Documents/chatgpt/node-chatgpt-api/node_modules/@fortaine/fetch-event-source/lib/esm/fetch.js:54:23)
at processTicksAndRejections (node:internal/process/task_queues:96:5)
Type "!" to access the command menu.
? Write a message: `

this is a bug? or my setting is error?

export default { // Your OpenAI API key openaiApiKey: 'apikey' || '', chatGptClient: { // (Optional) Parameters as described in https://platform.openai.com/docs/api-reference/completions modelOptions: { // The model is set to text-chat-davinci-002-20221122 by default, but you can override // it and any other parameters here //model: 'text-chat-davinci-002-20221122', model: 'text-davinci-003', }, // (Optional) Set custom instructions instead of "You are ChatGPT...". // promptPrefix: 'You are Bob, a cowboy in Western times...', // (Optional) Set a custom name for the user // userLabel: 'User', // (Optional) Set a custom name for ChatGPT // chatGptLabel: 'ChatGPT', // (Optional) Set to true to enable console.debug()logging debug: false, }, // Options for the Keyv cache, see https://www.npmjs.com/package/keyv. // This is used for storing conversations, and supports additional drivers (conversations are stored in memory by default) cacheOptions: {}, // Options for the API server apiOptions: { port: process.env.API_PORT || 3000, host: process.env.API_HOST || 'localhost', // (Optional) Set to true to enableconsole.debug()logging debug: false, }, // If set, ChatGPTClient will usekeyv-fileto store conversations to this JSON file instead of in memory. // However,cacheOptions.store will override this if set storageFilePath: process.env.STORAGE_FILE_PATH || './cache.json', }

My account has been removed from the waiting list, but running use-bing-client.js returns the following:

node .\use-bing-client.js

performing handshake

handshake established

{"arguments":[{"source":"cib","optionsSets":["nlu_direct_response_filter","deepleo","enable_debug_commands","disable_emoji_spoken_text","responsible_ai_policy_235","enablemm"],"isStartOfSession":true,"message":{"author":"user","inputMethod":"Keyboard","text":"Write a short poem about cats","messageType":"Chat"},"participant":{}}],"invocationId":"0","target":"chat","type":4}
2. --------------------------
[{"type":2,"invocationId":"0","item":{"firstNewMessageIndex":null,"suggestedResponses":null,"conversationId":"7104db73-d7ed-4428-baa3-3ab8f406e816","requestId":"4c647392-28de-4bcb-b0ff-c9a30be92cca","telemetry":{"metrics":null,"startTime":"2023-02-12T04:07:52.1888796Z"},"result":{"value":"UnauthorizedRequest","message":"The conversationSignature is not set and is required to ensure that you have permission to use our APIs. IsAuthenticated=False. App ID=","error":"UnauthorizedRequest","renewCert":true,"serviceVersion":"20230210.12"}}},{"type":3,"invocationId":"0"}]

Ability to set OpenAI API key for server on the first request to it

TypeError: fetch failed
at Object.fetch (node:internal/deps/undici/undici:11118:11)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async create (file:///app/node_modules/@waylaidwanderer/fetch-event-source/lib/esm/fetch.js:52:34) {
cause: Error: read ECONNRESET
at TLSWrap.onStreamRead (node:internal/stream_base_commons:217:20) {
errno: -104,
code: 'ECONNRESET',
syscall: 'read'
}
}

On Mac, when running cli server locally with npx chatgpt-cli results in error message:

env: node\r: No such file or directory

Apparently the problem is the CRLF EOL in cli.js not being recognized by the bash interpreter on macOS as described here. By converting CRLF to LF manually, the script works fine.

Request
{ "message": "I need a python script to check for balanced parenthesis in a string", "conversationId": "your-conversation-id (optional)", "parentMessageId": "your-parent-message-id (optional)" }

Response
{ "response": "```\ndef is_balanced(string):\n stack = []\n for char in string:\n if char in ('(', '{', '['):\n stack.append(char)\n elif char in (')', '}', ']'):\n if not stack:\n return False\n else:\n top = stack.pop()\n if top == '(' and char != ')':\n return False\n elif top == '{' and char != '}':\n return False\n elif top == '[' and char != ']':\n return False\n return not stack\n```", "conversationId": "your-conversation-id (optional)", "messageId": "0bfcc43d-c4a7-4822-9692-26b44b4de193" }

Please. Is there a good way to use it?

using reverse proxy
proxy: https://chatgpt.pawan.krd/api/completions
module: text-davinci-002-render
api_key: https://chat.openai.com/api/auth/session's accessToken

Questions

How stable is this for you in general? How many accts are you typically using and how many reqs / hour?

• can you DM me on discord what you're using for residential IP proxy? just curious

Feedback

• it'd be nice to have the pool functionality available as a module usable outside of this server (stretch goal; not necessary; just a thought)
• right now, the way you're initializing accts, it's hard to know when they're available after one or more initSession calls succeeds
• checking the response for special response strings is important for robustness
• maybe we could just port https://github.com/transitive-bullshit/chatgpt-twitter-bot/blob/main/src/chatgpt-api-pool.ts? I know you asked me to do this awhile ago, but I never got around to it..
• maybe add support for 2captcha as well as nopecha
• it looks you're periodically running refreshSession but never resetSession. does this work for you consistently in practice? ChatGPTAPIBrowser.sendMessage has some error handling & retry logic, but over time it's gotten pretty messy btwn the logic I have in there vs in the ChatGPTAPIPool in my twitter bot. Lots of stuff is duplicated, and it's kind of a mess.
• maybe add an example curl or httpie example to the readme to show how you're invoking the local API? something people can copy & paste
• conversation ID persistence will be important

Comments

• I really like the ability to use a proxy per acct

Sorry for the long feedback. I just want to make sure if I'm going to prominently direct people to this project from the chatgpt-api repo, that it encapsulates the robustness learnings I've come across from running the twitter bot.

The more feedback I write, the more I think this might not be useful feedback. I dunno; just stream of consciousness.

Cheers && thanks! 🙏 💕 🙏

can u give some sample how to work it ?

'That model is currently overloaded with other requests. You can retry your request, or contact us through our help center at help.openai.com if the error persists. (Please include the request ID de79efb84259b34394b36c1f27bb548c in your message.)',
type: 'server_error'

Hello,
I would like to get authorization to access and use the BingAIClient since I got the necessary "-U" cookie.

Could you explain to me how I can get authorization to access the project?

Message:
"result":{"value":"UnauthorizedRequest","message":"The conversationSignature is not set and is required to ensure that you have permission to use our APIs. IsAuthenticated=False. App ID=","error": "UnauthorizedRequest"

thanks in advance for your help.

when i test it ,it show that :

Error: Failed to send message. HTTP 503 - {
  "error": {
    "message": "That model is currently overloaded with other requests. You can retry your request, or contact us through our help center at help.openai.com if the error persists. (Ple
ase include the request ID 169bee8d  in your message.)",
    "type": "server_error",
    "param": null,
    "code": null
  }
}

fyi

Hi, sorry if this is a simple question.

I have a JS code to get "_U" cookie from a page. But I don't know how to use it on bing.com.

The blow is my funciton.

function getCookie() {
  const cookieName = "_U";
  const cookies = document.cookie.split(';');
  for (let i = 0; i < cookies.length; i++) {
    const cookie = cookies[i].trim();
    if (cookie.startsWith(cookieName + '=')) {
      return decodeURIComponent(cookie.substring(cookieName.length + 1));
    }
  }
  return null;
}

status: 404,
json: {
error: {
message: 'That model does not exist',
type: 'invalid_request_error',
param: null,
code: null
}

This is a good reference and I know it works on both ARM64 and AMD64: https://github.com/matrixgpt/matrix-chatgpt-bot/blob/main/Dockerfile

got this: I am an AI language model created by OpenAI, I am not ChatGPT, seems text-chat-davinci-002-20230126 is not ChatGPT model?

message: 'That model does not exist',
type: 'invalid_request_error',
param: null,
code: null

Hello. Looks like the model has been removed.

I use 3 accounts

But now I have a problem when I start a conversation and I get a conversationId and I want to use it, I get an error "Conversation not found" because the system created the conversation with Accoount1 and is now using Account 3

what is the best way to solve this?

what are the limits in bing?

nodejs 16.18.0
after set the apikey
npm install
npm start
when post to /conversation , the server echo en error

FetchError: request to https://api.openai.com/v1/completions failed, reason: getaddrinfo ENOTFOUND api.openai.com
at ClientRequest. (file:///D:/codes/node-chatgpt-api-1.15.0/node_modules/node-fetch/src/index.js:108:11)
at ClientRequest.emit (node:events:525:35)
at TLSSocket.socketErrorListener (node:_http_client:494:9)
at TLSSocket.emit (node:events:513:28)
at emitErrorNT (node:internal/streams/destroy:157:8)
at emitErrorCloseNT (node:internal/streams/destroy:122:3)
at processTicksAndRejections (node:internal/process/task_queues:83:21) {
type: 'system',
errno: 'ENOTFOUND',
code: 'ENOTFOUND',
erroredSysCall: 'getaddrinfo'
}

file:///C:/Users/Daniel/Documents/2023/OpenAi/node_modules/@waylaidwanderer/chatgpt-api/src/ChatGPTClient.js:47
const error = new Error(Failed to send message. HTTP ${response.status} - ${body});
^

Error: Failed to send message. HTTP 404 - {
"error": {
"message": "That model does not exist",
"type": "invalid_request_error",
"param": null,
"code": null
}
}

at ChatGPTClient.getCompletion (file:///C:/Users/Daniel/Documents/2023/OpenAi/node_modules/@waylaidwanderer/chatgpt-api/src/ChatGPTClient.js:47:27)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async ChatGPTClient.sendMessage (file:///C:/Users/Daniel/Documents/2023/OpenAi/node_modules/@waylaidwanderer/chatgpt-api/src/ChatGPTClient.js:83:24)
at async promptCompletion (file:///C:/Users/Daniel/Documents/2023/OpenAi/src/controllers/openAi/open.js:12:24) {

status: 404,
json: {
error: {
message: 'That model does not exist',
type: 'invalid_request_error',
param: null,
code: null
}
}
}