LibreChat/api/server/services/Endpoints/openAI/initialize.js
Danny Avila 37964975c1
🤖 refactor: Improve Agents Memory Usage, Bump Keyv, Grok 3 (#6850)
* chore: remove unused redis file

* chore: bump keyv dependencies, and update related imports

* refactor: Implement IoRedis client for rate limiting across middleware, as node-redis via keyv not compatible

* fix: Set max listeners to expected amount

* WIP: memory improvements

* refactor: Simplify getAbortData assignment in createAbortController

* refactor: Update getAbortData to use WeakRef for content management

* WIP: memory improvements in agent chat requests

* refactor: Enhance memory management with finalization registry and cleanup functions

* refactor: Simplify domainParser calls by removing unnecessary request parameter

* refactor: Update parameter types for action tools and agent loading functions to use minimal configs

* refactor: Simplify domainParser tests by removing unnecessary request parameter

* refactor: Simplify domainParser call by removing unnecessary request parameter

* refactor: Enhance client disposal by nullifying additional properties to improve memory management

* refactor: Improve title generation by adding abort controller and timeout handling, consolidate request cleanup

* refactor: Update checkIdleConnections to skip current user when checking for idle connections if passed

* refactor: Update createMCPTool to derive userId from config and handle abort signals

* refactor: Introduce createTokenCounter function and update tokenCounter usage; enhance disposeClient to reset Graph values

* refactor: Update getMCPManager to accept userId parameter for improved idle connection handling

* refactor: Extract logToolError function for improved error handling in AgentClient

* refactor: Update disposeClient to clear handlerRegistry and graphRunnable references in client.run

* refactor: Extract createHandleNewToken function to streamline token handling in initializeClient

* chore: bump @librechat/agents

* refactor: Improve timeout handling in addTitle function for better error management

* refactor: Introduce createFetch instead of using class method

* refactor: Enhance client disposal and request data handling in AskController and EditController

* refactor: Update import statements for AnthropicClient and OpenAIClient to use specific paths

* refactor: Use WeakRef for response handling in SplitStreamHandler to prevent memory leaks

* refactor: Simplify client disposal and rename getReqData to processReqData in AskController and EditController

* refactor: Improve logging structure and parameter handling in OpenAIClient

* refactor: Remove unused GraphEvents and improve stream event handling in AnthropicClient and OpenAIClient

* refactor: Simplify client initialization in AskController and EditController

* refactor: Remove unused mock functions and implement in-memory store for KeyvMongo

* chore: Update dependencies in package-lock.json to latest versions

* refactor: Await token usage recording in OpenAIClient to ensure proper async handling

* refactor: Remove handleAbort route from multiple endpoints and enhance client disposal logic

* refactor: Enhance abort controller logic by managing abortKey more effectively

* refactor: Add newConversation handling in useEventHandlers for improved conversation management

* fix: dropparams

* refactor: Use optional chaining for safer access to request properties in BaseClient

* refactor: Move client disposal and request data processing logic to cleanup module for better organization

* refactor: Remove aborted request check from addTitle function for cleaner logic

* feat: Add Grok 3 model pricing and update tests for new models

* chore: Remove trace warnings and inspect flags from backend start script used for debugging

* refactor: Replace user identifier handling with userId for consistency across controllers, use UserId in clientRegistry

* refactor: Enhance client disposal logic to prevent memory leaks by clearing additional references

* chore: Update @librechat/agents to version 2.4.14 in package.json and package-lock.json
2025-04-12 18:46:36 -04:00

168 lines
5 KiB
JavaScript

const {
ErrorTypes,
EModelEndpoint,
resolveHeaders,
mapModelToAzureConfig,
} = require('librechat-data-provider');
const { getUserKeyValues, checkUserKeyExpiry } = require('~/server/services/UserService');
const { getLLMConfig } = require('~/server/services/Endpoints/openAI/llm');
const { isEnabled, isUserProvided, sleep } = require('~/server/utils');
const OpenAIClient = require('~/app/clients/OpenAIClient');
const { getAzureCredentials } = require('~/utils');
function createHandleNewToken(streamRate) {
async () => {
await sleep(streamRate);
};
}
const initializeClient = async ({
req,
res,
endpointOption,
optionsOnly,
overrideEndpoint,
overrideModel,
}) => {
const {
PROXY,
OPENAI_API_KEY,
AZURE_API_KEY,
OPENAI_REVERSE_PROXY,
AZURE_OPENAI_BASEURL,
OPENAI_SUMMARIZE,
DEBUG_OPENAI,
} = process.env;
const { key: expiresAt } = req.body;
const modelName = overrideModel ?? req.body.model;
const endpoint = overrideEndpoint ?? req.body.endpoint;
const contextStrategy = isEnabled(OPENAI_SUMMARIZE) ? 'summarize' : null;
const credentials = {
[EModelEndpoint.openAI]: OPENAI_API_KEY,
[EModelEndpoint.azureOpenAI]: AZURE_API_KEY,
};
const baseURLOptions = {
[EModelEndpoint.openAI]: OPENAI_REVERSE_PROXY,
[EModelEndpoint.azureOpenAI]: AZURE_OPENAI_BASEURL,
};
const userProvidesKey = isUserProvided(credentials[endpoint]);
const userProvidesURL = isUserProvided(baseURLOptions[endpoint]);
let userValues = null;
if (expiresAt && (userProvidesKey || userProvidesURL)) {
checkUserKeyExpiry(expiresAt, endpoint);
userValues = await getUserKeyValues({ userId: req.user.id, name: endpoint });
}
let apiKey = userProvidesKey ? userValues?.apiKey : credentials[endpoint];
let baseURL = userProvidesURL ? userValues?.baseURL : baseURLOptions[endpoint];
let clientOptions = {
contextStrategy,
proxy: PROXY ?? null,
debug: isEnabled(DEBUG_OPENAI),
reverseProxyUrl: baseURL ? baseURL : null,
...endpointOption,
};
const isAzureOpenAI = endpoint === EModelEndpoint.azureOpenAI;
/** @type {false | TAzureConfig} */
const azureConfig = isAzureOpenAI && req.app.locals[EModelEndpoint.azureOpenAI];
if (isAzureOpenAI && azureConfig) {
const { modelGroupMap, groupMap } = azureConfig;
const {
azureOptions,
baseURL,
headers = {},
serverless,
} = mapModelToAzureConfig({
modelName,
modelGroupMap,
groupMap,
});
clientOptions.reverseProxyUrl = baseURL ?? clientOptions.reverseProxyUrl;
clientOptions.headers = resolveHeaders({ ...headers, ...(clientOptions.headers ?? {}) });
clientOptions.titleConvo = azureConfig.titleConvo;
clientOptions.titleModel = azureConfig.titleModel;
const azureRate = modelName.includes('gpt-4') ? 30 : 17;
clientOptions.streamRate = azureConfig.streamRate ?? azureRate;
clientOptions.titleMethod = azureConfig.titleMethod ?? 'completion';
const groupName = modelGroupMap[modelName].group;
clientOptions.addParams = azureConfig.groupMap[groupName].addParams;
clientOptions.dropParams = azureConfig.groupMap[groupName].dropParams;
clientOptions.forcePrompt = azureConfig.groupMap[groupName].forcePrompt;
apiKey = azureOptions.azureOpenAIApiKey;
clientOptions.azure = !serverless && azureOptions;
if (serverless === true) {
clientOptions.defaultQuery = azureOptions.azureOpenAIApiVersion
? { 'api-version': azureOptions.azureOpenAIApiVersion }
: undefined;
clientOptions.headers['api-key'] = apiKey;
}
} else if (isAzureOpenAI) {
clientOptions.azure = userProvidesKey ? JSON.parse(userValues.apiKey) : getAzureCredentials();
apiKey = clientOptions.azure.azureOpenAIApiKey;
}
/** @type {undefined | TBaseEndpoint} */
const openAIConfig = req.app.locals[EModelEndpoint.openAI];
if (!isAzureOpenAI && openAIConfig) {
clientOptions.streamRate = openAIConfig.streamRate;
clientOptions.titleModel = openAIConfig.titleModel;
}
/** @type {undefined | TBaseEndpoint} */
const allConfig = req.app.locals.all;
if (allConfig) {
clientOptions.streamRate = allConfig.streamRate;
}
if (userProvidesKey & !apiKey) {
throw new Error(
JSON.stringify({
type: ErrorTypes.NO_USER_KEY,
}),
);
}
if (!apiKey) {
throw new Error(`${endpoint} API Key not provided.`);
}
if (optionsOnly) {
const modelOptions = endpointOption.model_parameters;
modelOptions.model = modelName;
clientOptions = Object.assign({ modelOptions }, clientOptions);
clientOptions.modelOptions.user = req.user.id;
const options = getLLMConfig(apiKey, clientOptions);
const streamRate = clientOptions.streamRate;
if (!streamRate) {
return options;
}
options.llmConfig.callbacks = [
{
handleLLMNewToken: createHandleNewToken(streamRate),
},
];
return options;
}
const client = new OpenAIClient(apiKey, Object.assign({ req, res }, clientOptions));
return {
client,
openAIApiKey: apiKey,
};
};
module.exports = initializeClient;