mirror of
https://github.com/ikechan8370/chatgpt-plugin.git
synced 2025-12-16 13:27:08 +00:00
395 lines
14 KiB
TypeScript
395 lines
14 KiB
TypeScript
import Keyv from 'keyv'
|
|
import pTimeout from 'p-timeout'
|
|
import QuickLRU from 'quick-lru'
|
|
import { v4 as uuidv4 } from 'uuid'
|
|
|
|
import * as tokenizer from './tokenizer'
|
|
import * as types from './types'
|
|
import globalFetch from 'node-fetch'
|
|
import {qwen, Role} from "./types";
|
|
import {openai} from "../openai/types";
|
|
|
|
const CHATGPT_MODEL = 'qwen-turbo' // qwen-plus
|
|
|
|
const USER_LABEL_DEFAULT = 'User'
|
|
const ASSISTANT_LABEL_DEFAULT = '通义千问'
|
|
|
|
export class QwenApi {
|
|
protected _apiKey: string
|
|
protected _apiBaseUrl: string
|
|
protected _debug: boolean
|
|
|
|
protected _systemMessage: string
|
|
protected _completionParams: Omit<
|
|
types.qwen.CreateChatCompletionRequest,
|
|
'messages' | 'n'
|
|
>
|
|
protected _maxModelTokens: number
|
|
protected _maxResponseTokens: number
|
|
protected _fetch: types.FetchFn
|
|
|
|
protected _getMessageById: types.GetMessageByIdFunction
|
|
protected _upsertMessage: types.UpsertMessageFunction
|
|
|
|
protected _messageStore: Keyv<types.ChatMessage>
|
|
|
|
/**
|
|
* Creates a new client wrapper around Qwen's chat completion API, mimicing the official ChatGPT webapp's functionality as closely as possible.
|
|
*
|
|
* @param opts
|
|
*/
|
|
constructor(opts: types.QWenAPIOptions) {
|
|
const {
|
|
apiKey,
|
|
apiBaseUrl = 'https://dashscope.aliyuncs.com/api/v1',
|
|
debug = false,
|
|
messageStore,
|
|
completionParams,
|
|
parameters,
|
|
systemMessage,
|
|
getMessageById,
|
|
upsertMessage,
|
|
fetch = globalFetch
|
|
} = opts
|
|
|
|
this._apiKey = apiKey
|
|
this._apiBaseUrl = apiBaseUrl
|
|
this._debug = !!debug
|
|
this._fetch = fetch
|
|
|
|
this._completionParams = {
|
|
model: CHATGPT_MODEL,
|
|
parameters: {
|
|
top_p: 0.5,
|
|
top_k: 50,
|
|
temperature: 1.0,
|
|
seed: 114514,
|
|
enable_search: true,
|
|
result_format: "message",
|
|
incremental_output: false,
|
|
...parameters
|
|
},
|
|
...completionParams
|
|
}
|
|
|
|
this._systemMessage = systemMessage
|
|
|
|
if (this._systemMessage === undefined) {
|
|
const currentDate = new Date().toISOString().split('T')[0]
|
|
this._systemMessage = `You are Qwen, a large language model trained by Alibaba Cloud. Answer as concisely as possible.\nCurrent date: ${currentDate}`
|
|
}
|
|
|
|
this._getMessageById = getMessageById ?? this._defaultGetMessageById
|
|
this._upsertMessage = upsertMessage ?? this._defaultUpsertMessage
|
|
|
|
if (messageStore) {
|
|
this._messageStore = messageStore
|
|
} else {
|
|
this._messageStore = new Keyv<types.ChatMessage, any>({
|
|
store: new QuickLRU<string, types.ChatMessage>({ maxSize: 10000 })
|
|
})
|
|
}
|
|
|
|
if (!this._apiKey) {
|
|
throw new Error('Qwen missing required apiKey')
|
|
}
|
|
|
|
if (!this._fetch) {
|
|
throw new Error('Invalid environment; fetch is not defined')
|
|
}
|
|
|
|
if (typeof this._fetch !== 'function') {
|
|
throw new Error('Invalid "fetch" is not a function')
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Sends a message to the Qwen chat completions endpoint, waits for the response
|
|
* to resolve, and returns the response.
|
|
*
|
|
* If you want your response to have historical context, you must provide a valid `parentMessageId`.
|
|
*
|
|
* If you want to receive a stream of partial responses, use `opts.onProgress`.
|
|
*
|
|
* Set `debug: true` in the `ChatGPTAPI` constructor to log more info on the full prompt sent to the Qwen chat completions API. You can override the `systemMessage` in `opts` to customize the assistant's instructions.
|
|
*
|
|
* @param message - The prompt message to send
|
|
* @param opts.parentMessageId - Optional ID of the previous message in the conversation (defaults to `undefined`)
|
|
* @param opts.conversationId - Optional ID of the conversation (defaults to `undefined`)
|
|
* @param opts.messageId - Optional ID of the message to send (defaults to a random UUID)
|
|
* @param opts.systemMessage - Optional override for the chat "system message" which acts as instructions to the model (defaults to the ChatGPT system message)
|
|
* @param opts.timeoutMs - Optional timeout in milliseconds (defaults to no timeout)
|
|
* @param opts.onProgress - Optional callback which will be invoked every time the partial response is updated
|
|
* @param opts.abortSignal - Optional callback used to abort the underlying `fetch` call using an [AbortController](https://developer.mozilla.org/en-US/docs/Web/API/AbortController)
|
|
* @param opts.completionParams - Optional overrides to send to the [Qwen chat completion API](https://platform.openai.com/docs/api-reference/chat/create). Options like `temperature` and `presence_penalty` can be tweaked to change the personality of the assistant.
|
|
*
|
|
* @returns The response from ChatGPT
|
|
*/
|
|
async sendMessage(
|
|
text: string,
|
|
opts: types.SendMessageOptions = {},
|
|
role: Role = 'user',
|
|
): Promise<types.ChatMessage> {
|
|
let {
|
|
parentMessageId,
|
|
messageId = uuidv4(),
|
|
timeoutMs,
|
|
completionParams,
|
|
conversationId
|
|
} = opts
|
|
|
|
let { abortSignal } = opts
|
|
|
|
let abortController: AbortController = null
|
|
if (timeoutMs && !abortSignal) {
|
|
abortController = new AbortController()
|
|
abortSignal = abortController.signal
|
|
}
|
|
|
|
const message: types.ChatMessage = {
|
|
role,
|
|
id: messageId,
|
|
conversationId,
|
|
parentMessageId,
|
|
text,
|
|
}
|
|
|
|
const latestQuestion = message
|
|
|
|
let parameters = Object.assign(
|
|
this._completionParams.parameters,
|
|
completionParams.parameters
|
|
)
|
|
completionParams = Object.assign(this._completionParams, completionParams)
|
|
completionParams.parameters = parameters
|
|
const { messages, maxTokens, numTokens } = await this._buildMessages(
|
|
text,
|
|
role,
|
|
opts,
|
|
completionParams
|
|
)
|
|
|
|
console.log(`maxTokens: ${maxTokens}, numTokens: ${numTokens}`)
|
|
const result: types.ChatMessage & { conversation: qwen.ChatCompletionRequestMessage[] } = {
|
|
role: 'assistant',
|
|
id: uuidv4(),
|
|
conversationId,
|
|
parentMessageId: messageId,
|
|
text: undefined,
|
|
functionCall: undefined,
|
|
conversation: []
|
|
}
|
|
completionParams.input = { messages }
|
|
const responseP = new Promise<types.ChatMessage>(
|
|
async (resolve, reject) => {
|
|
const url = `${this._apiBaseUrl}/services/aigc/text-generation/generation`
|
|
const headers = {
|
|
'Content-Type': 'application/json',
|
|
Authorization: `Bearer ${this._apiKey}`
|
|
}
|
|
const body = completionParams
|
|
if (this._debug) {
|
|
console.log(JSON.stringify(body))
|
|
}
|
|
|
|
if (this._debug) {
|
|
console.log(`sendMessage (${numTokens} tokens)`, body)
|
|
}
|
|
try {
|
|
const res = await this._fetch(url, {
|
|
method: 'POST',
|
|
headers,
|
|
body: JSON.stringify(body),
|
|
signal: abortSignal
|
|
})
|
|
|
|
if (!res.ok) {
|
|
const reason = await res.text()
|
|
const msg = `Qwen error ${
|
|
res.status || res.statusText
|
|
}: ${reason}`
|
|
const error = new types.ChatGPTError(msg, { cause: res })
|
|
error.statusCode = res.status
|
|
error.statusText = res.statusText
|
|
return reject(error)
|
|
}
|
|
|
|
const response: types.qwen.CreateChatCompletionResponse =
|
|
await res.json()
|
|
if (this._debug) {
|
|
console.log(response)
|
|
}
|
|
if (response.output?.choices?.[0]?.message?.tool_calls?.length > 0) {
|
|
// function call result
|
|
result.functionCall = response.output.choices[0].message.tool_calls[0].function
|
|
}
|
|
if (response?.request_id) {
|
|
result.id = response.request_id
|
|
}
|
|
result.detail = response
|
|
result.text = response.output.choices[0].message.content
|
|
result.conversation = messages
|
|
return resolve(result)
|
|
} catch (err) {
|
|
return reject(err)
|
|
}
|
|
|
|
}
|
|
).then(async (message) => {
|
|
return Promise.all([
|
|
this._upsertMessage(latestQuestion),
|
|
this._upsertMessage(message)
|
|
]).then(() => message)
|
|
})
|
|
|
|
if (timeoutMs) {
|
|
if (abortController) {
|
|
// This will be called when a timeout occurs in order for us to forcibly
|
|
// ensure that the underlying HTTP request is aborted.
|
|
;(responseP as any).cancel = () => {
|
|
abortController.abort()
|
|
}
|
|
}
|
|
|
|
return pTimeout(responseP, {
|
|
milliseconds: timeoutMs,
|
|
message: 'Qwen timed out waiting for response'
|
|
})
|
|
} else {
|
|
return responseP
|
|
}
|
|
}
|
|
|
|
get apiKey(): string {
|
|
return this._apiKey
|
|
}
|
|
|
|
set apiKey(apiKey: string) {
|
|
this._apiKey = apiKey
|
|
}
|
|
|
|
|
|
protected async _buildMessages(text: string, role: Role, opts: types.SendMessageOptions, completionParams: Partial<
|
|
Omit<qwen.CreateChatCompletionRequest, 'messages' | 'n' | 'stream'>
|
|
>) {
|
|
const { systemMessage = this._systemMessage } = opts
|
|
let { parentMessageId } = opts
|
|
|
|
const userLabel = USER_LABEL_DEFAULT
|
|
const assistantLabel = ASSISTANT_LABEL_DEFAULT
|
|
|
|
// fix number of qwen
|
|
const maxNumTokens = 6000
|
|
let messages: types.qwen.ChatCompletionRequestMessage[] = []
|
|
|
|
if (systemMessage) {
|
|
messages.push({
|
|
role: 'system',
|
|
content: systemMessage
|
|
})
|
|
}
|
|
|
|
const systemMessageOffset = messages.length
|
|
let nextMessages = text
|
|
? messages.concat([
|
|
{
|
|
role,
|
|
content: text,
|
|
name: role === 'tool' ? opts.name : undefined
|
|
}
|
|
])
|
|
: messages
|
|
|
|
let functionToken = 0
|
|
|
|
let numTokens = functionToken
|
|
|
|
do {
|
|
const prompt = nextMessages
|
|
.reduce((prompt, message) => {
|
|
switch (message.role) {
|
|
case 'system':
|
|
return prompt.concat([`Instructions:\n${message.content}`])
|
|
case 'user':
|
|
return prompt.concat([`${userLabel}:\n${message.content}`])
|
|
default:
|
|
return message.content ? prompt.concat([`${assistantLabel}:\n${message.content}`]) : prompt
|
|
}
|
|
}, [] as string[])
|
|
.join('\n\n')
|
|
|
|
let nextNumTokensEstimate = await this._getTokenCount(prompt)
|
|
|
|
for (const m1 of nextMessages) {
|
|
nextNumTokensEstimate += await this._getTokenCount('')
|
|
}
|
|
|
|
const isValidPrompt = nextNumTokensEstimate + functionToken <= maxNumTokens
|
|
|
|
if (prompt && !isValidPrompt) {
|
|
break
|
|
}
|
|
messages = nextMessages
|
|
numTokens = nextNumTokensEstimate + functionToken
|
|
|
|
if (!isValidPrompt) {
|
|
break
|
|
}
|
|
|
|
if (!parentMessageId) {
|
|
break
|
|
}
|
|
|
|
const parentMessage = await this._getMessageById(parentMessageId)
|
|
if (!parentMessage) {
|
|
break
|
|
}
|
|
|
|
const parentMessageRole = parentMessage.role || 'user'
|
|
|
|
nextMessages = nextMessages.slice(0, systemMessageOffset).concat([
|
|
{
|
|
role: parentMessageRole,
|
|
content: parentMessage.functionCall ? parentMessage.functionCall.arguments : parentMessage.text,
|
|
name: parentMessage.functionCall ? parentMessage.functionCall.name : undefined
|
|
},
|
|
...nextMessages.slice(systemMessageOffset)
|
|
])
|
|
|
|
parentMessageId = parentMessage.parentMessageId
|
|
|
|
} while (true)
|
|
|
|
// Use up to 4096 tokens (prompt + response), but try to leave 1000 tokens
|
|
// for the response.
|
|
const maxTokens = Math.max(
|
|
1,
|
|
Math.min(this._maxModelTokens - numTokens, this._maxResponseTokens)
|
|
)
|
|
|
|
return { messages, maxTokens, numTokens }
|
|
}
|
|
|
|
protected async _getTokenCount(text: string) {
|
|
if (!text) {
|
|
return 0
|
|
}
|
|
// TODO: use a better fix in the tokenizer
|
|
text = text.replace(/<\|endoftext\|>/g, '')
|
|
|
|
return tokenizer.encode(text).length
|
|
}
|
|
|
|
protected async _defaultGetMessageById(
|
|
id: string
|
|
): Promise<types.ChatMessage> {
|
|
const res = await this._messageStore.get(id)
|
|
return res
|
|
}
|
|
|
|
protected async _defaultUpsertMessage(
|
|
message: types.ChatMessage
|
|
): Promise<void> {
|
|
await this._messageStore.set(message.request_id, message)
|
|
}
|
|
}
|