🅰️ feat: Azure Config to Allow Different Deployments per Model (#1863)

* wip: first pass for azure endpoint schema

* refactor: azure config to return groupMap and modelConfigMap

* wip: naming and schema changes

* refactor(errorsToString): move to data-provider

* feat: rename to azureGroups, add additional tests, tests all expected outcomes, return errors

* feat(AppService): load Azure groups

* refactor(azure): use imported types, write `mapModelToAzureConfig`

* refactor: move `extractEnvVariable` to data-provider

* refactor(validateAzureGroups): throw on duplicate groups or models; feat(mapModelToAzureConfig): throw if env vars not present, add tests

* refactor(AppService): ensure each model is properly configured on startup

* refactor: deprecate azureOpenAI environment variables in favor of librechat.yaml config

* feat: use helper functions to handle and order enabled/default endpoints; initialize azureOpenAI from config file

* refactor: redefine types as well as load azureOpenAI models from config file

* chore(ci): fix test description naming

* feat(azureOpenAI): use validated model grouping for request authentication

* chore: bump data-provider following rebase

* chore: bump config file version noting significant changes

* feat: add title options and switch azure configs for titling and vision requests

* feat: enable azure plugins from config file

* fix(ci): pass tests

* chore(.env.example): mark `PLUGINS_USE_AZURE` as deprecated

* fix(fetchModels): early return if apiKey not passed

* chore: fix azure config typing

* refactor(mapModelToAzureConfig): return baseURL and headers as well as azureOptions

* feat(createLLM): use `azureOpenAIBasePath`

* feat(parsers): resolveHeaders

* refactor(extractBaseURL): handle invalid input

* feat(OpenAIClient): handle headers and baseURL for azureConfig

* fix(ci): pass `OpenAIClient` tests

* chore: extract env var for azureOpenAI group config, baseURL

* docs: azureOpenAI config setup docs

* feat: safe check of potential conflicting env vars that map to unique placeholders

* fix: reset apiKey when model switches from originally requested model (vision or title)

* chore: linting

* docs: CONFIG_PATH notes in custom_config.md

api/server/services/Config/loadAsyncEndpoints.js

@@ -1,12 +1,14 @@
-const { availableTools } = require('~/app/clients/tools');
+const { EModelEndpoint } = require('librechat-data-provider');
 const { addOpenAPISpecs } = require('~/app/clients/tools/util/addOpenAPISpecs');
+const { availableTools } = require('~/app/clients/tools');
 const { openAIApiKey, azureOpenAIApiKey, useAzurePlugins, userProvidedOpenAI, googleKey } =
   require('./EndpointService').config;
 
 /**
  * Load async endpoints and return a configuration object
+ * @param {Express.Request} req - The request object
  */
-async function loadAsyncEndpoints() {
+async function loadAsyncEndpoints(req) {
   let i = 0;
   let serviceKey, googleUserProvides;
   try {
@@ -35,13 +37,14 @@ async function loadAsyncEndpoints() {
 
   const google = serviceKey || googleKey ? { userProvide: googleUserProvides } : false;
 
+  const useAzure = req.app.locals[EModelEndpoint.azureOpenAI]?.plugins;
   const gptPlugins =
-    openAIApiKey || azureOpenAIApiKey
+    useAzure || openAIApiKey || azureOpenAIApiKey
       ? {
         plugins,
         availableAgents: ['classic', 'functions'],
-        userProvide: userProvidedOpenAI,
-        azure: useAzurePlugins,
+        userProvide: useAzure ? false : userProvidedOpenAI,
+        azure: useAzurePlugins || useAzure,
       }
       : false;
 

api/server/services/Config/loadConfigEndpoints.js

@@ -1,11 +1,13 @@
-const { EModelEndpoint } = require('librechat-data-provider');
-const { isUserProvided, extractEnvVariable } = require('~/server/utils');
+const { EModelEndpoint, extractEnvVariable } = require('librechat-data-provider');
+const { isUserProvided } = require('~/server/utils');
 const getCustomConfig = require('./getCustomConfig');
 
 /**
  * Load config endpoints from the cached configuration object
- * @function loadConfigEndpoints */
-async function loadConfigEndpoints() {
+ * @param {Express.Request} req - The request object
+ * @returns {Promise<TEndpointsConfig>} A promise that resolves to an object containing the endpoints configuration
+ */
+async function loadConfigEndpoints(req) {
   const customConfig = await getCustomConfig();
 
   if (!customConfig) {
@@ -42,6 +44,13 @@ async function loadConfigEndpoints() {
     }
   }
 
+  if (req.app.locals[EModelEndpoint.azureOpenAI]) {
+    /** @type {Omit<TConfig, 'order'>} */
+    endpointsConfig[EModelEndpoint.azureOpenAI] = {
+      userProvide: false,
+    };
+  }
+
   return endpointsConfig;
 }
 

api/server/services/Config/loadConfigModels.js

@@ -1,5 +1,5 @@
-const { EModelEndpoint } = require('librechat-data-provider');
-const { isUserProvided, extractEnvVariable } = require('~/server/utils');
+const { EModelEndpoint, extractEnvVariable } = require('librechat-data-provider');
+const { isUserProvided } = require('~/server/utils');
 const { fetchModels } = require('~/server/services/ModelService');
 const getCustomConfig = require('./getCustomConfig');
 
@@ -17,6 +17,16 @@ async function loadConfigModels(req) {
 
   const { endpoints = {} } = customConfig ?? {};
   const modelsConfig = {};
+  const azureModels = req.app.locals[EModelEndpoint.azureOpenAI]?.modelNames;
+  const azureEndpoint = endpoints[EModelEndpoint.azureOpenAI];
+
+  if (azureModels && azureEndpoint) {
+    modelsConfig[EModelEndpoint.azureOpenAI] = azureModels;
+  }
+
+  if (azureModels && azureEndpoint && azureEndpoint.plugins) {
+    modelsConfig[EModelEndpoint.gptPlugins] = azureModels;
+  }
 
   if (!Array.isArray(endpoints[EModelEndpoint.custom])) {
     return modelsConfig;

api/server/services/Config/loadCustomConfig.js

@@ -18,7 +18,7 @@ let i = 0;
 async function loadCustomConfig() {
   // Use CONFIG_PATH if set, otherwise fallback to defaultConfigPath
   const configPath = process.env.CONFIG_PATH || defaultConfigPath;
-  
+
   const customConfig = loadYaml(configPath);
   if (!customConfig) {
     i === 0 &&

api/server/services/Config/loadDefaultEConfig.js

@@ -1,34 +1,17 @@
-const { EModelEndpoint } = require('librechat-data-provider');
+const { EModelEndpoint, getEnabledEndpoints } = require('librechat-data-provider');
 const loadAsyncEndpoints = require('./loadAsyncEndpoints');
 const { config } = require('./EndpointService');
 
 /**
  * Load async endpoints and return a configuration object
- * @function loadDefaultEndpointsConfig
+ * @param {Express.Request} req - The request object
  * @returns {Promise<Object.<string, EndpointWithOrder>>} An object whose keys are endpoint names and values are objects that contain the endpoint configuration and an order.
  */
-async function loadDefaultEndpointsConfig() {
-  const { google, gptPlugins } = await loadAsyncEndpoints();
+async function loadDefaultEndpointsConfig(req) {
+  const { google, gptPlugins } = await loadAsyncEndpoints(req);
   const { openAI, assistants, bingAI, anthropic, azureOpenAI, chatGPTBrowser } = config;
 
-  let enabledEndpoints = [
-    EModelEndpoint.openAI,
-    EModelEndpoint.assistants,
-    EModelEndpoint.azureOpenAI,
-    EModelEndpoint.google,
-    EModelEndpoint.bingAI,
-    EModelEndpoint.chatGPTBrowser,
-    EModelEndpoint.gptPlugins,
-    EModelEndpoint.anthropic,
-  ];
-
-  const endpointsEnv = process.env.ENDPOINTS || '';
-  if (endpointsEnv) {
-    enabledEndpoints = endpointsEnv
-      .split(',')
-      .filter((endpoint) => endpoint?.trim())
-      .map((endpoint) => endpoint.trim());
-  }
+  const enabledEndpoints = getEnabledEndpoints();
 
   const endpointConfig = {
     [EModelEndpoint.openAI]: openAI,