🛠️ refactor: Model Loading and Custom Endpoint Error Handling (#1849)

* fix: handle non-assistant role ChatCompletionMessage error

* refactor(ModelController): decouple res.send from loading/caching models

* fix(custom/initializeClient): only fetch custom endpoint models if models.fetch is true

* refactor(validateModel): load models if modelsConfig is not yet cached

* docs: update on file upload rate limiting

api/server/controllers/ModelController.js

@@ -2,12 +2,16 @@ const { CacheKeys } = require('librechat-data-provider');
 const { loadDefaultModels, loadConfigModels } = require('~/server/services/Config');
 const { getLogStores } = require('~/cache');
 
-async function modelController(req, res) {
+/**
+ * Loads the models from the config.
+ * @param {Express.Request} req - The Express request object.
+ * @returns {Promise<TModelsConfig>} The models config.
+ */
+async function loadModels(req) {
   const cache = getLogStores(CacheKeys.CONFIG_STORE);
   const cachedModelsConfig = await cache.get(CacheKeys.MODELS_CONFIG);
   if (cachedModelsConfig) {
-    res.send(cachedModelsConfig);
-    return;
+    return cachedModelsConfig;
   }
   const defaultModelsConfig = await loadDefaultModels(req);
   const customModelsConfig = await loadConfigModels(req);
@@ -15,7 +19,12 @@ async function modelController(req, res) {
   const modelConfig = { ...defaultModelsConfig, ...customModelsConfig };
 
   await cache.set(CacheKeys.MODELS_CONFIG, modelConfig);
+  return modelConfig;
+}
+
+async function modelController(req, res) {
+  const modelConfig = await loadModels(req);
   res.send(modelConfig);
 }
 
-module.exports = modelController;
+module.exports = { modelController, loadModels };

api/server/middleware/validateModel.js

@@ -1,4 +1,5 @@
 const { EModelEndpoint, CacheKeys, ViolationTypes } = require('librechat-data-provider');
+const { loadModels } = require('~/server/controllers/ModelController');
 const { logViolation, getLogStores } = require('~/cache');
 const { handleError } = require('~/server/utils');
 
@@ -17,7 +18,11 @@ const validateModel = async (req, res, next) => {
   }
 
   const cache = getLogStores(CacheKeys.CONFIG_STORE);
-  const modelsConfig = await cache.get(CacheKeys.MODELS_CONFIG);
+  let modelsConfig = await cache.get(CacheKeys.MODELS_CONFIG);
+  if (!modelsConfig) {
+    modelsConfig = await loadModels(req);
+  }
+
   if (!modelsConfig) {
     return handleError(res, { text: 'Models not loaded' });
   }

api/server/routes/models.js

@@ -1,8 +1,8 @@
 const express = require('express');
-const router = express.Router();
-const controller = require('../controllers/ModelController');
-const { requireJwtAuth } = require('../middleware/');
+const { modelController } = require('~/server/controllers/ModelController');
+const { requireJwtAuth } = require('~/server/middleware/');
 
-router.get('/', requireJwtAuth, controller);
+const router = express.Router();
+router.get('/', requireJwtAuth, modelController);
 
 module.exports = router;

api/server/services/Endpoints/custom/initializeClient.js

@@ -41,7 +41,7 @@ const initializeClient = async ({ req, res, endpointOption }) => {
 
   const cache = getLogStores(CacheKeys.TOKEN_CONFIG);
   let endpointTokenConfig = await cache.get(endpoint);
-  if (!endpointTokenConfig) {
+  if (endpointConfig && endpointConfig.models.fetch && !endpointTokenConfig) {
     await fetchModels({ apiKey: CUSTOM_API_KEY, baseURL: CUSTOM_BASE_URL, name: endpoint });
     endpointTokenConfig = await cache.get(endpoint);
   }