Merge pull request #1 from danny-avila/main

Updates from their main branch

.devcontainer/Dockerfile

@@ -0,0 +1,5 @@
+FROM node:18-bullseye
+
+RUN useradd -m -s /bin/bash vscode
+RUN mkdir -p /workspaces && chown -R vscode:vscode /workspaces
+WORKDIR /workspaces

.devcontainer/devcontainer.json

@@ -13,5 +13,6 @@
     }
   },
   "postCreateCommand": "",
-  "features": { "ghcr.io/devcontainers/features/git:1": {} }
+  "features": { "ghcr.io/devcontainers/features/git:1": {} },
+  "remoteUser": "vscode"
 }

.devcontainer/docker-compose.yml

@@ -2,7 +2,9 @@ version: "3.8"
 
 services:
   app:
-    image: node:19-bullseye
+    build:
+      context: ..
+      dockerfile: .devcontainer/Dockerfile
     # restart: always
     links:
       - mongodb
@@ -30,8 +32,8 @@ services:
     # Use "forwardPorts" in **devcontainer.json** to forward an app port locally. 
     # (Adding the "ports" property to this file will not forward from a Codespace.)
 
-    # Uncomment the next line to use a non-root user for all processes - See https://aka.ms/vscode-remote/containers/non-root for details.
-    # user: vscode
+    # Use a non-root user for all processes - See https://aka.ms/vscode-remote/containers/non-root for details.
+    user: vscode
 
     # Overrides default command so things don't shut down after the process ends.
     command: /bin/sh -c "while sleep 1000; do :; done"  

.env.example

@@ -1,13 +1,13 @@
-#=============================================================#
-#                   LibreChat Configuration                   #
-#=============================================================#
-# Please refer to the reference documentation for assistance  #
-# with configuring your LibreChat environment. The guide is   #
-# available both online and within your local LibreChat       #
-# directory:                                                  #
-# Online: https://docs.librechat.ai/install/dotenv.html       #
-# Locally: ./docs/install/dotenv.md                           #
-#=============================================================#
+#=====================================================================#
+#                       LibreChat Configuration                       #
+#=====================================================================#
+# Please refer to the reference documentation for assistance          #
+# with configuring your LibreChat environment. The guide is           #
+# available both online and within your local LibreChat               #
+# directory:                                                          #
+# Online: https://docs.librechat.ai/install/configuration/dotenv.html #
+# Locally: ./docs/install/configuration/dotenv.md                     #
+#=====================================================================#
 
 #==================================================#
 #               Server Configuration               #
@@ -86,7 +86,7 @@ BINGAI_TOKEN=user_provided
 
 CHATGPT_TOKEN=
 CHATGPT_MODELS=text-davinci-002-render-sha
-# CHATGPT_REVERSE_PROXY=<YOUR REVERSE PROXY>
+# CHATGPT_REVERSE_PROXY=
 
 #============#
 # Google     #
@@ -101,7 +101,7 @@ GOOGLE_KEY=user_provided
 #============#
 
 OPENAI_API_KEY=user_provided
-# OPENAI_MODELS=gpt-3.5-turbo-1106,gpt-4-1106-preview,gpt-3.5-turbo,gpt-3.5-turbo-16k,gpt-3.5-turbo-0301,gpt-4,gpt-4-0314,gpt-4-0613
+# OPENAI_MODELS=gpt-3.5-turbo-0125,gpt-3.5-turbo-0301,gpt-3.5-turbo,gpt-4,gpt-4-0613,gpt-4-vision-preview,gpt-3.5-turbo-0613,gpt-3.5-turbo-16k-0613,gpt-4-0125-preview,gpt-4-turbo-preview,gpt-4-1106-preview,gpt-3.5-turbo-1106,gpt-3.5-turbo-instruct,gpt-3.5-turbo-instruct-0914,gpt-3.5-turbo-16k
 
 DEBUG_OPENAI=false
 
@@ -127,7 +127,7 @@ DEBUG_OPENAI=false
 # Plugins    #
 #============#
 
-# PLUGIN_MODELS=gpt-3.5-turbo,gpt-3.5-turbo-16k,gpt-3.5-turbo-0301,gpt-4,gpt-4-0314,gpt-4-0613
+# PLUGIN_MODELS=gpt-4,gpt-4-turbo-preview,gpt-4-0125-preview,gpt-4-1106-preview,gpt-4-0613,gpt-3.5-turbo,gpt-3.5-turbo-0125,gpt-3.5-turbo-1106,gpt-3.5-turbo-0613
 
 DEBUG_PLUGINS=true
 
@@ -147,20 +147,20 @@ AZURE_AI_SEARCH_SEARCH_OPTION_SELECT=
 
 # DALL·E
 #----------------
-# DALLE_API_KEY= # Key for both DALL-E-2 and DALL-E-3
-# DALLE3_API_KEY= # Key for DALL-E-3 only
-# DALLE2_API_KEY= # Key for DALL-E-2 only
-# DALLE3_SYSTEM_PROMPT="Your DALL-E-3 System Prompt here"
-# DALLE2_SYSTEM_PROMPT="Your DALL-E-2 System Prompt here"
-# DALLE_REVERSE_PROXY= # Reverse proxy for DALL-E-2 and DALL-E-3
-# DALLE3_BASEURL= # Base URL for DALL-E-3
-# DALLE2_BASEURL= # Base URL for DALL-E-2
+# DALLE_API_KEY=
+# DALLE3_API_KEY=
+# DALLE2_API_KEY=
+# DALLE3_SYSTEM_PROMPT=
+# DALLE2_SYSTEM_PROMPT=
+# DALLE_REVERSE_PROXY=
+# DALLE3_BASEURL=
+# DALLE2_BASEURL=
 
 # DALL·E (via Azure OpenAI)
 # Note: requires some of the variables above to be set
 #----------------
-# DALLE3_AZURE_API_VERSION= # Azure OpenAI API version for DALL-E-3
-# DALLE2_AZURE_API_VERSION= # Azure OpenAI API versiion for DALL-E-2
+# DALLE3_AZURE_API_VERSION=
+# DALLE2_AZURE_API_VERSION=
 
 # Google
 #-----------------
@@ -202,7 +202,7 @@ MEILI_MASTER_KEY=DrhYf7zENyR6AlUCKmnz0eYASOQdl6zxH7s7MKFSfFCt
 
 OPENAI_MODERATION=false
 OPENAI_MODERATION_API_KEY=
-# OPENAI_MODERATION_REVERSE_PROXY=not working with some reverse proxys
+# OPENAI_MODERATION_REVERSE_PROXY=
 
 BAN_VIOLATIONS=true
 BAN_DURATION=1000 * 60 * 60 * 2

README.md

@@ -27,15 +27,15 @@
 </p>
 
 <p align="center">
-  <a href="https://railway.app/template/b5k2mn?referralCode=HI9hWz">
-    <img src="https://railway.app/button.svg" alt="Deploy on Railway">
-  </a>
-</p>
-
-<p align="center">
-  <a href="https://template.cloud.sealos.io/deploy?templateName=librechat">
-    <img src="https://raw.githubusercontent.com/labring-actions/templates/main/Deploy-on-Sealos.svg" alt="Deploy on Sealos">
-  </a>
+<a href="https://railway.app/template/b5k2mn?referralCode=HI9hWz">
+  <img src="https://railway.app/button.svg" alt="Deploy on Railway" height="30">
+</a>
+<a href="https://zeabur.com/templates/0X2ZY8">
+  <img src="https://zeabur.com/button.svg" alt="Deploy on Zeabur" height="30"/>
+</a>
+<a href="https://template.cloud.sealos.io/deploy?templateName=librechat">
+  <img src="https://raw.githubusercontent.com/labring-actions/templates/main/Deploy-on-Sealos.svg" alt="Deploy on Sealos" height="30">
+</a>
 </p>
 
 # 📃 Features

api/app/clients/BaseClient.js

@@ -428,7 +428,10 @@ class BaseClient {
       await this.saveMessageToDatabase(userMessage, saveOptions, user);
     }
 
-    if (isEnabled(process.env.CHECK_BALANCE) && supportsBalanceCheck[this.options.endpoint]) {
+    if (
+      isEnabled(process.env.CHECK_BALANCE) &&
+      supportsBalanceCheck[this.options.endpointType ?? this.options.endpoint]
+    ) {
       await checkBalance({
         req: this.options.req,
         res: this.options.res,
@@ -438,6 +441,7 @@ class BaseClient {
           amount: promptTokens,
           model: this.modelOptions.model,
           endpoint: this.options.endpoint,
+          endpointTokenConfig: this.options.endpointTokenConfig,
         },
       });
     }

api/app/clients/OpenAIClient.js

@@ -131,7 +131,13 @@ class OpenAIClient extends BaseClient {
     const { isChatGptModel } = this;
     this.isUnofficialChatGptModel =
       model.startsWith('text-chat') || model.startsWith('text-davinci-002-render');
-    this.maxContextTokens = getModelMaxTokens(model) ?? 4095; // 1 less than maximum
+
+    this.maxContextTokens =
+      getModelMaxTokens(
+        model,
+        this.options.endpointType ?? this.options.endpoint,
+        this.options.endpointTokenConfig,
+      ) ?? 4095; // 1 less than maximum
 
     if (this.shouldSummarize) {
       this.maxContextTokens = Math.floor(this.maxContextTokens / 2);
@@ -779,7 +785,12 @@ ${convo}
     // TODO: remove the gpt fallback and make it specific to endpoint
     const { OPENAI_SUMMARY_MODEL = 'gpt-3.5-turbo' } = process.env ?? {};
     const model = this.options.summaryModel ?? OPENAI_SUMMARY_MODEL;
-    const maxContextTokens = getModelMaxTokens(model) ?? 4095;
+    const maxContextTokens =
+      getModelMaxTokens(
+        model,
+        this.options.endpointType ?? this.options.endpoint,
+        this.options.endpointTokenConfig,
+      ) ?? 4095; // 1 less than maximum
 
     // 3 tokens for the assistant label, and 98 for the summarizer prompt (101)
     let promptBuffer = 101;
@@ -885,6 +896,7 @@ ${convo}
         model: this.modelOptions.model,
         context: 'message',
         conversationId: this.conversationId,
+        endpointTokenConfig: this.options.endpointTokenConfig,
       },
       { promptTokens, completionTokens },
     );
@@ -975,9 +987,22 @@ ${convo}
         ...opts,
       });
 
-      /* hacky fix for Mistral AI API not allowing a singular system message in payload */
+      /* hacky fixes for Mistral AI API:
+      - Re-orders system message to the top of the messages payload, as not allowed anywhere else
+      - If there is only one message and it's a system message, change the role to user
+      */
       if (opts.baseURL.includes('https://api.mistral.ai/v1') && modelOptions.messages) {
         const { messages } = modelOptions;
+
+        const systemMessageIndex = messages.findIndex((msg) => msg.role === 'system');
+
+        if (systemMessageIndex > 0) {
+          const [systemMessage] = messages.splice(systemMessageIndex, 1);
+          messages.unshift(systemMessage);
+        }
+
+        modelOptions.messages = messages;
+
         if (messages.length === 1 && messages[0].role === 'system') {
           modelOptions.messages[0].role = 'user';
         }

api/app/clients/tools/util/handleTools.test.js

@@ -53,6 +53,7 @@ describe('Tool Handlers', () => {
       username: 'fakeuser',
       email: '[email protected]',
       emailVerified: false,
+      // file deepcode ignore NoHardcodedPasswords/test: fake value
       password: 'fakepassword123',
       avatar: '',
       provider: 'local',

api/cache/getLogStores.js

@@ -23,6 +23,18 @@ const config = isEnabled(USE_REDIS)
   ? new Keyv({ store: keyvRedis })
   : new Keyv({ namespace: CacheKeys.CONFIG_STORE });
 
+const tokenConfig = isEnabled(USE_REDIS) // ttl: 30 minutes
+  ? new Keyv({ store: keyvRedis, ttl: 1800000 })
+  : new Keyv({ namespace: CacheKeys.TOKEN_CONFIG, ttl: 1800000 });
+
+const genTitle = isEnabled(USE_REDIS) // ttl: 2 minutes
+  ? new Keyv({ store: keyvRedis, ttl: 120000 })
+  : new Keyv({ namespace: CacheKeys.GEN_TITLE, ttl: 120000 });
+
+const modelQueries = isEnabled(process.env.USE_REDIS)
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: 'models' });
+
 const namespaces = {
   [CacheKeys.CONFIG_STORE]: config,
   pending_req,
@@ -34,6 +46,9 @@ const namespaces = {
   token_balance: createViolationInstance('token_balance'),
   registrations: createViolationInstance('registrations'),
   logins: createViolationInstance('logins'),
+  [CacheKeys.TOKEN_CONFIG]: tokenConfig,
+  [CacheKeys.GEN_TITLE]: genTitle,
+  [CacheKeys.MODEL_QUERIES]: modelQueries,
 };
 
 /**

api/models/Balance.js

@@ -10,8 +10,9 @@ balanceSchema.statics.check = async function ({
   valueKey,
   tokenType,
   amount,
+  endpointTokenConfig,
 }) {
-  const multiplier = getMultiplier({ valueKey, tokenType, model, endpoint });
+  const multiplier = getMultiplier({ valueKey, tokenType, model, endpoint, endpointTokenConfig });
   const tokenCost = amount * multiplier;
   const { tokenCredits: balance } = (await this.findOne({ user }, 'tokenCredits').lean()) ?? {};
 
@@ -24,6 +25,7 @@ balanceSchema.statics.check = async function ({
     amount,
     balance,
     multiplier,
+    endpointTokenConfig: !!endpointTokenConfig,
   });
 
   if (!balance) {

api/models/Conversation.js

@@ -30,12 +30,12 @@ module.exports = {
       return { message: 'Error saving conversation' };
     }
   },
-  getConvosByPage: async (user, pageNumber = 1, pageSize = 14) => {
+  getConvosByPage: async (user, pageNumber = 1, pageSize = 25) => {
     try {
       const totalConvos = (await Conversation.countDocuments({ user })) || 1;
       const totalPages = Math.ceil(totalConvos / pageSize);
       const convos = await Conversation.find({ user })
-        .sort({ createdAt: -1 })
+        .sort({ updatedAt: -1 })
         .skip((pageNumber - 1) * pageSize)
         .limit(pageSize)
         .lean();
@@ -45,7 +45,7 @@ module.exports = {
       return { message: 'Error getting conversations' };
     }
   },
-  getConvosQueried: async (user, convoIds, pageNumber = 1, pageSize = 14) => {
+  getConvosQueried: async (user, convoIds, pageNumber = 1, pageSize = 25) => {
     try {
       if (!convoIds || convoIds.length === 0) {
         return { conversations: [], pages: 1, pageNumber, pageSize };

api/models/Transaction.js

@@ -10,8 +10,8 @@ transactionSchema.methods.calculateTokenValue = function () {
   if (!this.valueKey || !this.tokenType) {
     this.tokenValue = this.rawAmount;
   }
-  const { valueKey, tokenType, model } = this;
-  const multiplier = getMultiplier({ valueKey, tokenType, model });
+  const { valueKey, tokenType, model, endpointTokenConfig } = this;
+  const multiplier = getMultiplier({ valueKey, tokenType, model, endpointTokenConfig });
   this.rate = multiplier;
   this.tokenValue = this.rawAmount * multiplier;
   if (this.context && this.tokenType === 'completion' && this.context === 'incomplete') {
@@ -25,6 +25,7 @@ transactionSchema.statics.create = async function (transactionData) {
   const Transaction = this;
 
   const transaction = new Transaction(transactionData);
+  transaction.endpointTokenConfig = transactionData.endpointTokenConfig;
   transaction.calculateTokenValue();
 
   // Save the transaction

api/models/checkBalance.js

@@ -14,6 +14,7 @@ const { logViolation } = require('../cache');
  * @param {('prompt' | 'completion')} params.txData.tokenType - The type of token.
  * @param {number} params.txData.amount - The amount of tokens.
  * @param {string} params.txData.model - The model name or identifier.
+ * @param {string} [params.txData.endpointTokenConfig] - The token configuration for the endpoint.
  * @returns {Promise<boolean>} Returns true if the user can spend the amount, otherwise denies the request.
  * @throws {Error} Throws an error if there's an issue with the balance check.
  */

api/models/schema/convoSchema.js

@@ -55,7 +55,7 @@ if (process.env.MEILI_HOST && process.env.MEILI_MASTER_KEY) {
   });
 }
 
-convoSchema.index({ createdAt: 1 });
+convoSchema.index({ createdAt: 1, updatedAt: 1 });
 
 const Conversation = mongoose.models.Conversation || mongoose.model('Conversation', convoSchema);
 

api/models/spendTokens.js

@@ -11,6 +11,7 @@ const { logger } = require('~/config');
  * @param {String} txData.conversationId - The ID of the conversation.
  * @param {String} txData.model - The model name.
  * @param {String} txData.context - The context in which the transaction is made.
+ * @param {String} [txData.endpointTokenConfig] - The current endpoint token config.
  * @param {String} [txData.valueKey] - The value key (optional).
  * @param {Object} tokenUsage - The number of tokens used.
  * @param {Number} tokenUsage.promptTokens - The number of prompt tokens used.

api/models/tx.js

@@ -12,6 +12,7 @@ const tokenValues = {
   '16k': { prompt: 3, completion: 4 },
   'gpt-3.5-turbo-1106': { prompt: 1, completion: 2 },
   'gpt-4-1106': { prompt: 10, completion: 30 },
+  'gpt-3.5-turbo-0125': { prompt: 0.5, completion: 1.5 },
 };
 
 /**
@@ -29,12 +30,18 @@ const getValueKey = (model, endpoint) => {
 
   if (modelName.includes('gpt-3.5-turbo-16k')) {
     return '16k';
+  } else if (modelName.includes('gpt-3.5-turbo-0125')) {
+    return 'gpt-3.5-turbo-0125';
   } else if (modelName.includes('gpt-3.5-turbo-1106')) {
     return 'gpt-3.5-turbo-1106';
   } else if (modelName.includes('gpt-3.5')) {
     return '4k';
   } else if (modelName.includes('gpt-4-1106')) {
     return 'gpt-4-1106';
+  } else if (modelName.includes('gpt-4-0125')) {
+    return 'gpt-4-1106';
+  } else if (modelName.includes('gpt-4-turbo')) {
+    return 'gpt-4-1106';
   } else if (modelName.includes('gpt-4-32k')) {
     return '32k';
   } else if (modelName.includes('gpt-4')) {
@@ -53,9 +60,14 @@ const getValueKey = (model, endpoint) => {
  * @param {string} [params.tokenType] - The type of token (e.g., 'prompt' or 'completion').
  * @param {string} [params.model] - The model name to derive the value key from if not provided.
  * @param {string} [params.endpoint] - The endpoint name to derive the value key from if not provided.
+ * @param {EndpointTokenConfig} [params.endpointTokenConfig] - The token configuration for the endpoint.
  * @returns {number} The multiplier for the given parameters, or a default value if not found.
  */
-const getMultiplier = ({ valueKey, tokenType, model, endpoint }) => {
+const getMultiplier = ({ valueKey, tokenType, model, endpoint, endpointTokenConfig }) => {
+  if (endpointTokenConfig) {
+    return endpointTokenConfig?.[model]?.[tokenType] ?? defaultRate;
+  }
+
   if (valueKey && tokenType) {
     return tokenValues[valueKey][tokenType] ?? defaultRate;
   }