✨ v0.7.3 (#3067)

* refactor: revert BaseClient to use node-fetch instead of undici

* chore: bump version

* chore: update npm dependencies to latest versions

* chore: fix custom footer

.env.example

@@ -2,11 +2,9 @@
 #                       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                     #
+# with configuring your LibreChat environment.                        #
+#                                                                     #
+# https://www.librechat.ai/docs/configuration/dotenv                  #
 #=====================================================================#
 
 #==================================================#
@@ -62,12 +60,15 @@ PROXY=
 #===================================#
 # Known Endpoints - librechat.yaml  #
 #===================================#
-# https://docs.librechat.ai/install/configuration/ai_endpoints.html
+# https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints
 
 # ANYSCALE_API_KEY=
 # APIPIE_API_KEY=
+# COHERE_API_KEY=
+# DATABRICKS_API_KEY=
 # FIREWORKS_API_KEY=
 # GROQ_API_KEY=
+# HUGGINGFACE_TOKEN=
 # MISTRAL_API_KEY=
 # OPENROUTER_KEY=
 # PERPLEXITY_API_KEY=
@@ -117,17 +118,31 @@ GOOGLE_KEY=user_provided
 # GOOGLE_REVERSE_PROXY=
 
 # Gemini API
-# GOOGLE_MODELS=gemini-1.0-pro,gemini-1.0-pro-001,gemini-1.0-pro-latest,gemini-1.0-pro-vision-latest,gemini-1.5-pro-latest,gemini-pro,gemini-pro-vision
+# GOOGLE_MODELS=gemini-1.5-flash-latest,gemini-1.0-pro,gemini-1.0-pro-001,gemini-1.0-pro-latest,gemini-1.0-pro-vision-latest,gemini-1.5-pro-latest,gemini-pro,gemini-pro-vision
 
 # Vertex AI
-# GOOGLE_MODELS=gemini-1.5-pro-preview-0409,gemini-1.0-pro-vision-001,gemini-pro,gemini-pro-vision,chat-bison,chat-bison-32k,codechat-bison,codechat-bison-32k,text-bison,text-bison-32k,text-unicorn,code-gecko,code-bison,code-bison-32k
+# GOOGLE_MODELS=gemini-1.5-flash-preview-0514,gemini-1.5-pro-preview-0514,gemini-1.0-pro-vision-001,gemini-1.0-pro-002,gemini-1.0-pro-001,gemini-pro-vision,gemini-1.0-pro
+
+# Google Gemini Safety Settings
+# NOTE (Vertex AI): You do not have access to the BLOCK_NONE setting by default.
+# To use this restricted HarmBlockThreshold setting, you will need to either:
+#
+# (a) Get access through an allowlist via your Google account team
+# (b) Switch your account type to monthly invoiced billing following this instruction:
+#     https://cloud.google.com/billing/docs/how-to/invoiced-billing
+#
+# GOOGLE_SAFETY_SEXUALLY_EXPLICIT=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_HATE_SPEECH=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_HARASSMENT=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_DANGEROUS_CONTENT=BLOCK_ONLY_HIGH
+
 
 #============#
 # OpenAI     #
 #============#
 
 OPENAI_API_KEY=user_provided
-# 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
+# OPENAI_MODELS=gpt-4o,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
 
@@ -149,7 +164,17 @@ DEBUG_OPENAI=false
 
 ASSISTANTS_API_KEY=user_provided
 # ASSISTANTS_BASE_URL=
-# ASSISTANTS_MODELS=gpt-3.5-turbo-0125,gpt-3.5-turbo-16k-0613,gpt-3.5-turbo-16k,gpt-3.5-turbo,gpt-4,gpt-4-0314,gpt-4-32k-0314,gpt-4-0613,gpt-3.5-turbo-0613,gpt-3.5-turbo-1106,gpt-4-0125-preview,gpt-4-turbo-preview,gpt-4-1106-preview
+# ASSISTANTS_MODELS=gpt-4o,gpt-3.5-turbo-0125,gpt-3.5-turbo-16k-0613,gpt-3.5-turbo-16k,gpt-3.5-turbo,gpt-4,gpt-4-0314,gpt-4-32k-0314,gpt-4-0613,gpt-3.5-turbo-0613,gpt-3.5-turbo-1106,gpt-4-0125-preview,gpt-4-turbo-preview,gpt-4-1106-preview
+
+#==========================#
+#   Azure Assistants API   #
+#==========================#
+
+# Note: You should map your credentials with custom variables according to your Azure OpenAI Configuration
+# The models for Azure Assistants are also determined by your Azure OpenAI configuration.
+
+# More info, including how to enable use of Assistants with Azure here:
+# https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints/azure#using-assistants-with-azure
 
 #============#
 # OpenRouter #
@@ -161,7 +186,7 @@ ASSISTANTS_API_KEY=user_provided
 # Plugins    #
 #============#
 
-# 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
+# PLUGIN_MODELS=gpt-4o,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
 
@@ -234,6 +259,14 @@ MEILI_NO_ANALYTICS=true
 MEILI_HOST=http://0.0.0.0:7700
 MEILI_MASTER_KEY=DrhYf7zENyR6AlUCKmnz0eYASOQdl6zxH7s7MKFSfFCt
 
+
+#==================================================#
+#          Speech to Text & Text to Speech         #
+#==================================================#
+
+STT_API_KEY=
+TTS_API_KEY=
+
 #===================================================#
 #                    User System                    #
 #===================================================#
@@ -288,6 +321,9 @@ ALLOW_EMAIL_LOGIN=true
 ALLOW_REGISTRATION=true
 ALLOW_SOCIAL_LOGIN=false
 ALLOW_SOCIAL_REGISTRATION=false
+ALLOW_PASSWORD_RESET=false
+# ALLOW_ACCOUNT_DELETION=true # note: enabled by default if omitted/commented out
+ALLOW_UNVERIFIED_EMAIL_LOGIN=true
 
 SESSION_EXPIRY=1000 * 60 * 15
 REFRESH_TOKEN_EXPIRY=(1000 * 60 * 60 * 24) * 7
@@ -329,6 +365,14 @@ OPENID_REQUIRED_ROLE_PARAMETER_PATH=
 OPENID_BUTTON_LABEL=
 OPENID_IMAGE_URL=
 
+# LDAP
+LDAP_URL=
+LDAP_BIND_DN=
+LDAP_BIND_CREDENTIALS=
+LDAP_USER_SEARCH_BASE=
+LDAP_SEARCH_FILTER=mail={{username}}
+LDAP_CA_CERT_PATH=
+
 #========================#
 # Email Password Reset   #
 #========================#
@@ -355,6 +399,13 @@ FIREBASE_STORAGE_BUCKET=
 FIREBASE_MESSAGING_SENDER_ID=
 FIREBASE_APP_ID=
 
+#========================#
+# Shared Links           #
+#========================#
+
+ALLOW_SHARED_LINKS=true
+ALLOW_SHARED_LINKS_PUBLIC=true
+
 #===================================================#
 #                        UI                         #
 #===================================================#
@@ -365,6 +416,9 @@ HELP_AND_FAQ_URL=https://librechat.ai
 
 # SHOW_BIRTHDAY_ICON=true
 
+# Google tag manager id
+#ANALYTICS_GTM_ID=user provided google tag manager id
+
 #==================================================#
 #                      Others                      #
 #==================================================#

.eslintrc.js

@@ -132,6 +132,13 @@ module.exports = {
         },
       ],
     },
+    {
+      files: './config/translations/**/*.ts',
+      parser: '@typescript-eslint/parser',
+      parserOptions: {
+        project: './config/translations/tsconfig.json',
+      },
+    },
     {
       files: ['./packages/data-provider/specs/**/*.ts'],
       parserOptions: {

.github/CONTRIBUTING.md

@@ -126,6 +126,18 @@ Apply the following naming conventions to branches, labels, and other Git-relate
       
    - **Current Stance**: At present, this backend transition is of lower priority and might not be pursued.
 
+## 7. Module Import Conventions
+
+- `npm` packages first, 
+     - from shortest line (top) to longest (bottom)
+
+- Followed by typescript types (pertains to data-provider and client workspaces)
+     - longest line (top) to shortest (bottom)
+     - types from package come first
+
+- Lastly, local imports
+     - longest line (top) to shortest (bottom)
+     - imports with alias `~` treated the same as relative import with respect to line length
 
 ---
 

.github/ISSUE_TEMPLATE/FEATURE-REQUEST.yml

@@ -43,7 +43,7 @@ body:
     id: terms
     attributes:
       label: Code of Conduct
-      description: By submitting this issue, you agree to follow our [Code of Conduct](https://github.com/danny-avila/LibreChat/blob/main/CODE_OF_CONDUCT.md)
+      description: By submitting this issue, you agree to follow our [Code of Conduct](https://github.com/danny-avila/LibreChat/blob/main/.github/CODE_OF_CONDUCT.md)
       options:
         - label: I agree to follow this project's Code of Conduct
           required: true

.github/ISSUE_TEMPLATE/QUESTION.yml

@@ -44,7 +44,7 @@ body:
     id: terms
     attributes:
       label: Code of Conduct
-      description: By submitting this issue, you agree to follow our [Code of Conduct](https://github.com/danny-avila/LibreChat/blob/main/CODE_OF_CONDUCT.md)
+      description: By submitting this issue, you agree to follow our [Code of Conduct](https://github.com/danny-avila/LibreChat/blob/main/.github/CODE_OF_CONDUCT.md)
       options:
         - label: I agree to follow this project's Code of Conduct
           required: true

.github/pull_request_template.md

@@ -1,7 +1,10 @@
 # Pull Request Template
 
+⚠️ Before Submitting a PR, Please Review:
+- Please ensure that you have thoroughly read and understood the [Contributing Docs](https://github.com/danny-avila/LibreChat/blob/main/.github/CONTRIBUTING.md) before submitting your Pull Request.
 
-### ⚠️ Before Submitting a PR, read the [Contributing Docs](https://github.com/danny-avila/LibreChat/blob/main/.github/CONTRIBUTING.md) in full!
+⚠️ Documentation Updates Notice:
+- Kindly note that documentation updates are managed in this repository: [librechat.ai](https://github.com/LibreChat-AI/librechat.ai)
 
 ## Summary
 
@@ -16,8 +19,6 @@ Please delete any irrelevant options.
 - [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
 - [ ] This change requires a documentation update
 - [ ] Translation update
-- [ ] Documentation update
-
 
 ## Testing
 
@@ -37,4 +38,4 @@ Please delete any irrelevant options.
 - [ ] I have written tests demonstrating that my changes are effective or that my feature works
 - [ ] Local unit tests pass with my changes
 - [ ] Any changes dependent on mine have been merged and published in downstream modules.
-- [ ] New documents have been locally validated with mkdocs
+- [ ] A pull request for updating the documentation has been submitted.

.github/workflows/frontend-review.yml

@@ -1,11 +1,6 @@
-#github action to run unit tests for frontend with jest
 name: Frontend Unit Tests
+
 on:
-  # push:
-  #   branches: 
-  #     - main
-  #     - dev
-  #     - release/*
   pull_request:
     branches: 
       - main
@@ -14,9 +9,10 @@ on:
     paths:
       - 'client/**'
       - 'packages/**'
+
 jobs:
-  tests_frontend:
-    name: Run frontend unit tests
+  tests_frontend_ubuntu:
+    name: Run frontend unit tests on Ubuntu
     timeout-minutes: 60
     runs-on: ubuntu-latest
     steps:
@@ -35,4 +31,26 @@ jobs:
 
       - name: Run unit tests
         run: npm run test:ci --verbose
-        working-directory: client
+        working-directory: client
+
+  tests_frontend_windows:
+    name: Run frontend unit tests on Windows
+    timeout-minutes: 60
+    runs-on: windows-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Use Node.js 20.x
+        uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: 'npm'
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Build Client
+        run: npm run frontend:ci
+
+      - name: Run unit tests
+        run: npm run test:ci --verbose
+        working-directory: client

.github/workflows/mkdocs.yaml

@@ -1,27 +0,0 @@
-name: mkdocs
-on:
-  push:
-    branches:
-      - main
-permissions:
-  contents: write
-jobs:
-  deploy:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v4
-        with:
-          python-version: 3.x
-      - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV 
-      - uses: actions/cache@v3
-        with:
-          key: mkdocs-material-${{ env.cache_id }}
-          path: .cache
-          restore-keys: |
-            mkdocs-material-
-      - run: pip install mkdocs-material 
-      - run: pip install mkdocs-nav-weight
-      - run: pip install mkdocs-publisher
-      - run: pip install mkdocs-exclude
-      - run: mkdocs gh-deploy --force

.gitignore

@@ -21,6 +21,10 @@ coverage
 # Grunt intermediate storage (http://gruntjs.com/creating-plugins#storing-task-files)
 .grunt
 
+# translation services
+config/translations/stores/*
+client/src/localization/languages/*_missing_keys.json
+
 # Compiled Dirs (http://nodejs.org/api/addons.html)
 build/
 dist/
@@ -69,6 +73,8 @@ src/style - official.css
 /playwright/.cache/
 .DS_Store
 *.code-workspace
+.idx
+monospace.json
 .idea
 *.iml
 *.pem

Dockerfile

@@ -1,10 +1,8 @@
-# v0.7.1
+# v0.7.3
 
 # Base node image
-FROM node:18-alpine3.18 AS node
+FROM node:20-alpine AS node
 
-RUN apk add g++ make py3-pip
-RUN npm install -g node-gyp
 RUN apk --no-cache add curl
 
 RUN mkdir -p /app && chown node:node /app
@@ -14,20 +12,20 @@ USER node
 
 COPY --chown=node:node . .
 
-# Allow mounting of these files, which have no default
-# values.
-RUN touch .env
-RUN npm config set fetch-retry-maxtimeout 600000
-RUN npm config set fetch-retries 5
-RUN npm config set fetch-retry-mintimeout 15000
-RUN npm install --no-audit
+RUN \
+    # Allow mounting of these files, which have no default
+    touch .env ; \
+    # Create directories for the volumes to inherit the correct permissions
+    mkdir -p /app/client/public/images /app/api/logs ; \
+    npm config set fetch-retry-maxtimeout 600000 ; \
+    npm config set fetch-retries 5 ; \
+    npm config set fetch-retry-mintimeout 15000 ; \
+    npm install --no-audit; \
+    # React client build
+    NODE_OPTIONS="--max-old-space-size=2048" npm run frontend; \
+    npm prune --production; \
+    npm cache clean --force
 
-# React client build
-ENV NODE_OPTIONS="--max-old-space-size=2048"
-RUN npm run frontend
-
-# Create directories for the volumes to inherit
-# the correct permissions
 RUN mkdir -p /app/client/public/images /app/api/logs
 
 # Node API setup

Dockerfile.multi

@@ -1,4 +1,4 @@
-# v0.7.1
+# v0.7.3
 
 # Build API, Client and Data Provider
 FROM node:20-alpine AS base
@@ -7,32 +7,31 @@ FROM node:20-alpine AS base
 FROM base AS data-provider-build
 WORKDIR /app/packages/data-provider
 COPY ./packages/data-provider ./
-RUN npm install
+RUN npm install; npm cache clean --force
 RUN npm run build
+RUN npm prune --production
 
 # React client build
-FROM data-provider-build AS client-build
+FROM base AS client-build
 WORKDIR /app/client
 COPY ./client/package*.json ./
 # Copy data-provider to client's node_modules
-RUN mkdir -p /app/client/node_modules/librechat-data-provider/
-RUN cp -R /app/packages/data-provider/* /app/client/node_modules/librechat-data-provider/
-RUN npm install
+COPY --from=data-provider-build /app/packages/data-provider/ /app/client/node_modules/librechat-data-provider/
+RUN npm install; npm cache clean --force
 COPY ./client/ ./
 ENV NODE_OPTIONS="--max-old-space-size=2048"
 RUN npm run build
 
 # Node API setup
-FROM data-provider-build AS api-build
+FROM base AS api-build
 WORKDIR /app/api
 COPY api/package*.json ./
 COPY api/ ./
 # Copy helper scripts
 COPY config/ ./
 # Copy data-provider to API's node_modules
-RUN mkdir -p /app/api/node_modules/librechat-data-provider/
-RUN cp -R /app/packages/data-provider/* /app/api/node_modules/librechat-data-provider/
-RUN npm install
+COPY --from=data-provider-build /app/packages/data-provider/ /app/api/node_modules/librechat-data-provider/
+RUN npm install --include prod; npm cache clean --force
 COPY --from=client-build /app/client/dist /app/client/dist
 EXPOSE 3080
 ENV HOST=0.0.0.0

README.md

@@ -1,6 +1,6 @@
 <p align="center">
   <a href="https://librechat.ai">
-    <img src="docs/assets/LibreChat.svg" height="256">
+    <img src="client/public/assets/logo.svg" height="256">
   </a>
   <h1 align="center">
     <a href="https://librechat.ai">LibreChat</a>
@@ -43,14 +43,14 @@
 - 🖥️ UI matching ChatGPT, including Dark mode, Streaming, and latest updates
 - 🤖 AI model selection:
   - OpenAI, Azure OpenAI, BingAI, ChatGPT, Google Vertex AI, Anthropic (Claude), Plugins, Assistants API (including Azure Assistants)
-- ✅ Compatible across both **[Remote & Local AI services](https://docs.librechat.ai/install/configuration/ai_endpoints.html#intro):**
+- ✅ Compatible across both **[Remote & Local AI services](https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints):**
   - groq, Ollama, Cohere, Mistral AI, Apple MLX, koboldcpp, OpenRouter, together.ai, Perplexity, ShuttleAI, and more
 - 💾 Create, Save, & Share Custom Presets
 - 🔀 Switch between AI Endpoints and Presets, mid-chat
 - 🔄 Edit, Resubmit, and Continue Messages with Conversation branching
 - 🌿 Fork Messages & Conversations for Advanced Context control
 - 💬 Multimodal Chat:
-    - Upload and analyze images with Claude 3, GPT-4, and Gemini Vision 📸
+    - Upload and analyze images with Claude 3, GPT-4 (including `gpt-4o`), and Gemini Vision 📸
     - Chat with Files using Custom Endpoints, OpenAI, Azure, Anthropic, & Google. 🗃️
     - Advanced Agents with Files, Code Interpreter, Tools, and API Actions 🔦
       - Available through the [OpenAI Assistants API](https://platform.openai.com/docs/assistants/overview) 🌤️
@@ -58,9 +58,13 @@
 - 🌎 Multilingual UI:
   - English, 中文, Deutsch, Español, Français, Italiano, Polski, Português Brasileiro,
   - Русский, 日本語, Svenska, 한국어, Tiếng Việt, 繁體中文, العربية, Türkçe, Nederlands, עברית
-- 🎨 Customizable Dropdown & Interface: Adapts to both power users and newcomers.
+- 🎨 Customizable Dropdown & Interface: Adapts to both power users and newcomers
+- 📧 Verify your email to ensure secure access
+- 🗣️ Chat hands-free with Speech-to-Text and Text-to-Speech magic
+  - Automatically send and play Audio
+  - Supports OpenAI, Azure OpenAI, and Elevenlabs
 - 📥 Import Conversations from LibreChat, ChatGPT, Chatbot UI
-- 📤 Export conversations as screenshots, markdown, text, json.
+- 📤 Export conversations as screenshots, markdown, text, json
 - 🔍 Search all messages/conversations
 - 🔌 Plugins, including web access, image generation with DALL-E-3 and more
 - 👥 Multi-User, Secure Authentication with Moderation and Token spend tools
@@ -69,7 +73,7 @@
 - 📖 Completely Open-Source & Built in Public
 - 🧑‍🤝‍🧑 Community-driven development, support, and feedback
 
-[For a thorough review of our features, see our docs here](https://docs.librechat.ai/features/plugins/introduction.html) 📚
+[For a thorough review of our features, see our docs here](https://docs.librechat.ai/) 📚
 
 ## 🪶 All-In-One AI Conversations with LibreChat
 
@@ -77,37 +81,49 @@ LibreChat brings together the future of assistant AIs with the revolutionary tec
 
 With LibreChat, you no longer need to opt for ChatGPT Plus and can instead use free or pay-per-call APIs. We welcome contributions, cloning, and forking to enhance the capabilities of this advanced chatbot platform.
 
-<!-- https://github.com/danny-avila/LibreChat/assets/110412045/c1eb0c0f-41f6-4335-b982-84b278b53d59 -->
-
-[![Watch the video](https://img.youtube.com/vi/pNIOs1ovsXw/maxresdefault.jpg)](https://youtu.be/pNIOs1ovsXw)
+[![Watch the video](https://img.youtube.com/vi/YLVUW5UP9N0/maxresdefault.jpg)](https://www.youtube.com/watch?v=YLVUW5UP9N0)
 Click on the thumbnail to open the video☝️
 
 ---
 
-## 📚 Documentation
+## 🌐 Resources
 
-For more information on how to use our advanced features, install and configure our software, and access our guidelines and tutorials, please check out our documentation at [docs.librechat.ai](https://docs.librechat.ai)
+**GitHub Repo:**
+  - **RAG API:** [github.com/danny-avila/rag_api](https://github.com/danny-avila/rag_api)
+  - **Website:** [github.com/LibreChat-AI/librechat.ai](https://github.com/LibreChat-AI/librechat.ai)
+
+**Other:**
+  - **Website:** [librechat.ai](https://librechat.ai)
+  - **Documentation:** [docs.librechat.ai](https://docs.librechat.ai)
+  - **Blog:** [blog.librechat.ai](https://docs.librechat.ai)
 
 ---
 
 ## 📝 Changelog
 
-Keep up with the latest updates by visiting the releases page - [Releases](https://github.com/danny-avila/LibreChat/releases)
+Keep up with the latest updates by visiting the releases page and notes:
+- [Releases](https://github.com/danny-avila/LibreChat/releases)
+- [Changelog](https://www.librechat.ai/changelog) 
 
-**⚠️ [Breaking Changes](docs/general_info/breaking_changes.md)**
-Please consult the breaking changes before updating.
+**⚠️ Please consult the [changelog](https://www.librechat.ai/changelog) for breaking changes before updating.**
 
 ---
 
 ## ⭐ Star History
 
 <p align="center">
-<a href="https://trendshift.io/repositories/4685" target="_blank"><img src="https://trendshift.io/api/badge/repositories/4685" alt="danny-avila%2FLibreChat | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+  <a href="https://star-history.com/#danny-avila/LibreChat&Date">
+    <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=danny-avila/LibreChat&type=Date&theme=dark" onerror="this.src='https://api.star-history.com/svg?repos=danny-avila/LibreChat&type=Date'" />
+  </a>
+</p>
+<p align="center">
+  <a href="https://trendshift.io/repositories/4685" target="_blank" style="padding: 10px;">
+    <img src="https://trendshift.io/api/badge/repositories/4685" alt="danny-avila%2FLibreChat | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/>
+  </a>
+  <a href="https://runacap.com/ross-index/q1-24/" target="_blank" rel="noopener" style="margin-left: 20px;">
+    <img style="width: 260px; height: 56px" src="https://runacap.com/wp-content/uploads/2024/04/ROSS_badge_white_Q1_2024.svg" alt="ROSS Index - Fastest Growing Open-Source Startups in Q1 2024 | Runa Capital" width="260" height="56"/>
+  </a>
 </p>
-
-<a href="https://star-history.com/#danny-avila/LibreChat&Date">
-  <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=danny-avila/LibreChat&type=Date&theme=dark" onerror="this.src='https://api.star-history.com/svg?repos=danny-avila/LibreChat&type=Date'" />
-</a>
 
 ---
 

api/app/clients/AnthropicClient.js

@@ -1,4 +1,5 @@
 const Anthropic = require('@anthropic-ai/sdk');
+const { HttpsProxyAgent } = require('https-proxy-agent');
 const { encoding_for_model: encodingForModel, get_encoding: getEncoding } = require('tiktoken');
 const {
   getResponseSender,
@@ -7,10 +8,10 @@ const {
 } = require('librechat-data-provider');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
 const {
-  titleFunctionPrompt,
-  parseTitleFromPrompt,
   truncateText,
   formatMessage,
+  titleFunctionPrompt,
+  parseParamFromPrompt,
   createContextHandlers,
 } = require('./prompts');
 const spendTokens = require('~/models/spendTokens');
@@ -75,7 +76,9 @@ class AnthropicClient extends BaseClient {
     this.options.attachments?.then((attachments) => this.checkVisionRequest(attachments));
 
     this.maxContextTokens =
-      getModelMaxTokens(this.modelOptions.model, EModelEndpoint.anthropic) ?? 100000;
+      this.options.maxContextTokens ??
+      getModelMaxTokens(this.modelOptions.model, EModelEndpoint.anthropic) ??
+      100000;
     this.maxResponseTokens = this.modelOptions.maxOutputTokens || 1500;
     this.maxPromptTokens =
       this.options.maxPromptTokens || this.maxContextTokens - this.maxResponseTokens;
@@ -121,9 +124,14 @@ class AnthropicClient extends BaseClient {
   getClient() {
     /** @type {Anthropic.default.RequestOptions} */
     const options = {
+      fetch: this.fetch,
       apiKey: this.apiKey,
     };
 
+    if (this.options.proxy) {
+      options.httpAgent = new HttpsProxyAgent(this.options.proxy);
+    }
+
     if (this.options.reverseProxyUrl) {
       options.baseURL = this.options.reverseProxyUrl;
     }
@@ -652,6 +660,7 @@ class AnthropicClient extends BaseClient {
 
   getSaveOptions() {
     return {
+      maxContextTokens: this.options.maxContextTokens,
       promptPrefix: this.options.promptPrefix,
       modelLabel: this.options.modelLabel,
       resendFiles: this.options.resendFiles,
@@ -745,7 +754,7 @@ class AnthropicClient extends BaseClient {
           context: 'title',
         });
         const text = response.content[0].text;
-        title = parseTitleFromPrompt(text);
+        title = parseParamFromPrompt(text, 'title');
       } catch (e) {
         logger.error('[AnthropicClient] There was an issue generating the title', e);
       }

api/app/clients/BaseClient.js

@@ -1,4 +1,5 @@
 const crypto = require('crypto');
+const fetch = require('node-fetch');
 const { supportsBalanceCheck, Constants } = require('librechat-data-provider');
 const { getConvo, getMessages, saveMessage, updateMessage, saveConvo } = require('~/models');
 const { addSpaceIfNeeded, isEnabled } = require('~/server/utils');
@@ -17,6 +18,7 @@ class BaseClient {
       month: 'long',
       day: 'numeric',
     });
+    this.fetch = this.fetch.bind(this);
   }
 
   setOptions() {
@@ -54,6 +56,25 @@ class BaseClient {
     });
   }
 
+  /**
+   * Makes an HTTP request and logs the process.
+   *
+   * @param {RequestInfo} url - The URL to make the request to. Can be a string or a Request object.
+   * @param {RequestInit} [init] - Optional init options for the request.
+   * @returns {Promise<Response>} - A promise that resolves to the response of the fetch request.
+   */
+  async fetch(_url, init) {
+    let url = _url;
+    if (this.options.directEndpoint) {
+      url = this.options.reverseProxyUrl;
+    }
+    logger.debug(`Making request to ${url}`);
+    if (typeof Bun !== 'undefined') {
+      return await fetch(url, init);
+    }
+    return await fetch(url, init);
+  }
+
   getBuildMessagesOptions() {
     throw new Error('Subclasses must implement getBuildMessagesOptions');
   }
@@ -373,6 +394,14 @@ class BaseClient {
     const { user, head, isEdited, conversationId, responseMessageId, saveOptions, userMessage } =
       await this.handleStartMethods(message, opts);
 
+    if (opts.progressCallback) {
+      opts.onProgress = opts.progressCallback.call(null, {
+        ...(opts.progressOptions ?? {}),
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
+    }
+
     const { generation = '' } = opts;
 
     // It's not necessary to push to currentMessages

api/app/clients/ChatGPTClient.js

@@ -438,9 +438,17 @@ class ChatGPTClient extends BaseClient {
 
       if (message.eventType === 'text-generation' && message.text) {
         onTokenProgress(message.text);
-      } else if (message.eventType === 'stream-end' && message.response) {
+        reply += message.text;
+      }
+      /*
+      Cohere API Chinese Unicode character replacement hotfix.
+      Should be un-commented when the following issue is resolved:
+      https://github.com/cohere-ai/cohere-typescript/issues/151
+
+      else if (message.eventType === 'stream-end' && message.response) {
         reply = message.response.text;
       }
+      */
     }
 
     return reply;

api/app/clients/GoogleClient.js

@@ -138,7 +138,10 @@ class GoogleClient extends BaseClient {
       !isGenerativeModel && !isChatModel && /code|text/.test(this.modelOptions.model);
     const { isTextModel } = this;
 
-    this.maxContextTokens = getModelMaxTokens(this.modelOptions.model, EModelEndpoint.google);
+    this.maxContextTokens =
+      this.options.maxContextTokens ??
+      getModelMaxTokens(this.modelOptions.model, EModelEndpoint.google);
+
     // The max prompt tokens is determined by the max context tokens minus the max response tokens.
     // Earlier messages will be dropped until the prompt is within the limit.
     this.maxResponseTokens = this.modelOptions.maxOutputTokens || settings.maxOutputTokens.default;
@@ -677,26 +680,36 @@ class GoogleClient extends BaseClient {
         };
       }
 
+      const safetySettings = _payload.safetySettings;
+      requestOptions.safetySettings = safetySettings;
+
+      const delay = modelName.includes('flash') ? 8 : 14;
       const result = await client.generateContentStream(requestOptions);
       for await (const chunk of result.stream) {
         const chunkText = chunk.text();
-        this.generateTextStream(chunkText, onProgress, {
-          delay: 12,
+        await this.generateTextStream(chunkText, onProgress, {
+          delay,
         });
         reply += chunkText;
       }
       return reply;
     }
 
+    const safetySettings = _payload.safetySettings;
     const stream = await model.stream(messages, {
       signal: abortController.signal,
       timeout: 7000,
+      safetySettings: safetySettings,
     });
 
+    let delay = this.isGenerativeModel ? 12 : 8;
+    if (modelName.includes('flash')) {
+      delay = 5;
+    }
     for await (const chunk of stream) {
       const chunkText = chunk?.content ?? chunk;
-      this.generateTextStream(chunkText, onProgress, {
-        delay: this.isGenerativeModel ? 12 : 8,
+      await this.generateTextStream(chunkText, onProgress, {
+        delay,
       });
       reply += chunkText;
     }
@@ -720,6 +733,33 @@ class GoogleClient extends BaseClient {
   }
 
   async sendCompletion(payload, opts = {}) {
+    const modelName = payload.parameters?.model;
+
+    if (modelName && modelName.toLowerCase().includes('gemini')) {
+      const safetySettings = [
+        {
+          category: 'HARM_CATEGORY_SEXUALLY_EXPLICIT',
+          threshold:
+            process.env.GOOGLE_SAFETY_SEXUALLY_EXPLICIT || 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
+        },
+        {
+          category: 'HARM_CATEGORY_HATE_SPEECH',
+          threshold: process.env.GOOGLE_SAFETY_HATE_SPEECH || 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
+        },
+        {
+          category: 'HARM_CATEGORY_HARASSMENT',
+          threshold: process.env.GOOGLE_SAFETY_HARASSMENT || 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
+        },
+        {
+          category: 'HARM_CATEGORY_DANGEROUS_CONTENT',
+          threshold:
+            process.env.GOOGLE_SAFETY_DANGEROUS_CONTENT || 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
+        },
+      ];
+
+      payload.safetySettings = safetySettings;
+    }
+
     let reply = '';
     reply = await this.getCompletion(payload, opts);
     return reply.trim();

api/app/clients/OllamaClient.js

@@ -0,0 +1,154 @@
+const { z } = require('zod');
+const axios = require('axios');
+const { Ollama } = require('ollama');
+const { deriveBaseURL } = require('~/utils');
+const { logger } = require('~/config');
+
+const ollamaPayloadSchema = z.object({
+  mirostat: z.number().optional(),
+  mirostat_eta: z.number().optional(),
+  mirostat_tau: z.number().optional(),
+  num_ctx: z.number().optional(),
+  repeat_last_n: z.number().optional(),
+  repeat_penalty: z.number().optional(),
+  temperature: z.number().optional(),
+  seed: z.number().nullable().optional(),
+  stop: z.array(z.string()).optional(),
+  tfs_z: z.number().optional(),
+  num_predict: z.number().optional(),
+  top_k: z.number().optional(),
+  top_p: z.number().optional(),
+  stream: z.optional(z.boolean()),
+  model: z.string(),
+});
+
+/**
+ * @param {string} imageUrl
+ * @returns {string}
+ * @throws {Error}
+ */
+const getValidBase64 = (imageUrl) => {
+  const parts = imageUrl.split(';base64,');
+
+  if (parts.length === 2) {
+    return parts[1];
+  } else {
+    logger.error('Invalid or no Base64 string found in URL.');
+  }
+};
+
+class OllamaClient {
+  constructor(options = {}) {
+    const host = deriveBaseURL(options.baseURL ?? 'http://localhost:11434');
+    /** @type {Ollama} */
+    this.client = new Ollama({ host });
+  }
+
+  /**
+   * Fetches Ollama models from the specified base API path.
+   * @param {string} baseURL
+   * @returns {Promise<string[]>} The Ollama models.
+   */
+  static async fetchModels(baseURL) {
+    let models = [];
+    if (!baseURL) {
+      return models;
+    }
+    try {
+      const ollamaEndpoint = deriveBaseURL(baseURL);
+      /** @type {Promise<AxiosResponse<OllamaListResponse>>} */
+      const response = await axios.get(`${ollamaEndpoint}/api/tags`);
+      models = response.data.models.map((tag) => tag.name);
+      return models;
+    } catch (error) {
+      const logMessage =
+        'Failed to fetch models from Ollama API. If you are not using Ollama directly, and instead, through some aggregator or reverse proxy that handles fetching via OpenAI spec, ensure the name of the endpoint doesn\'t start with `ollama` (case-insensitive).';
+      logger.error(logMessage, error);
+      return [];
+    }
+  }
+
+  /**
+   * @param {ChatCompletionMessage[]} messages
+   * @returns {OllamaMessage[]}
+   */
+  static formatOpenAIMessages(messages) {
+    const ollamaMessages = [];
+
+    for (const message of messages) {
+      if (typeof message.content === 'string') {
+        ollamaMessages.push({
+          role: message.role,
+          content: message.content,
+        });
+        continue;
+      }
+
+      let aggregatedText = '';
+      let imageUrls = [];
+
+      for (const content of message.content) {
+        if (content.type === 'text') {
+          aggregatedText += content.text + ' ';
+        } else if (content.type === 'image_url') {
+          imageUrls.push(getValidBase64(content.image_url.url));
+        }
+      }
+
+      const ollamaMessage = {
+        role: message.role,
+        content: aggregatedText.trim(),
+      };
+
+      if (imageUrls.length > 0) {
+        ollamaMessage.images = imageUrls;
+      }
+
+      ollamaMessages.push(ollamaMessage);
+    }
+
+    return ollamaMessages;
+  }
+
+  /***
+   * @param {Object} params
+   * @param {ChatCompletionPayload} params.payload
+   * @param {onTokenProgress} params.onProgress
+   * @param {AbortController} params.abortController
+   */
+  async chatCompletion({ payload, onProgress, abortController = null }) {
+    let intermediateReply = '';
+
+    const parameters = ollamaPayloadSchema.parse(payload);
+    const messages = OllamaClient.formatOpenAIMessages(payload.messages);
+
+    if (parameters.stream) {
+      const stream = await this.client.chat({
+        messages,
+        ...parameters,
+      });
+
+      for await (const chunk of stream) {
+        const token = chunk.message.content;
+        intermediateReply += token;
+        onProgress(token);
+        if (abortController.signal.aborted) {
+          stream.controller.abort();
+          break;
+        }
+      }
+    }
+    // TODO: regular completion
+    else {
+      // const generation = await this.client.generate(payload);
+    }
+
+    return intermediateReply;
+  }
+  catch(err) {
+    logger.error('[OllamaClient.chatCompletion]', err);
+    throw err;
+  }
+}
+
+module.exports = { OllamaClient, ollamaPayloadSchema };

api/app/clients/OpenAIClient.js

@@ -1,4 +1,5 @@
 const OpenAI = require('openai');
+const { OllamaClient } = require('./OllamaClient');
 const { HttpsProxyAgent } = require('https-proxy-agent');
 const {
   Constants,
@@ -26,8 +27,9 @@ const {
   createContextHandlers,
 } = require('./prompts');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
-const { handleOpenAIErrors } = require('./tools/util');
+const { updateTokenWebsocket } = require('~/server/services/Files/Audio');
 const { isEnabled, sleep } = require('~/server/utils');
+const { handleOpenAIErrors } = require('./tools/util');
 const spendTokens = require('~/models/spendTokens');
 const { createLLM, RunManager } = require('./llm');
 const ChatGPTClient = require('./ChatGPTClient');
@@ -128,6 +130,10 @@ class OpenAIClient extends BaseClient {
       this.useOpenRouter = true;
     }
 
+    if (this.options.endpoint?.toLowerCase() === 'ollama') {
+      this.isOllama = true;
+    }
+
     this.FORCE_PROMPT =
       isEnabled(OPENAI_FORCE_PROMPT) ||
       (reverseProxy && reverseProxy.includes('completions') && !reverseProxy.includes('chat'));
@@ -160,11 +166,13 @@ class OpenAIClient extends BaseClient {
       model.startsWith('text-chat') || model.startsWith('text-davinci-002-render');
 
     this.maxContextTokens =
+      this.options.maxContextTokens ??
       getModelMaxTokens(
         model,
         this.options.endpointType ?? this.options.endpoint,
         this.options.endpointTokenConfig,
-      ) ?? 4095; // 1 less than maximum
+      ) ??
+      4095; // 1 less than maximum
 
     if (this.shouldSummarize) {
       this.maxContextTokens = Math.floor(this.maxContextTokens / 2);
@@ -234,23 +242,52 @@ class OpenAIClient extends BaseClient {
    * @param {MongoFile[]} attachments
    */
   checkVisionRequest(attachments) {
-    const availableModels = this.options.modelsConfig?.[this.options.endpoint];
-    this.isVisionModel = validateVisionModel({ model: this.modelOptions.model, availableModels });
-
-    const visionModelAvailable = availableModels?.includes(this.defaultVisionModel);
-    if (
-      attachments &&
-      attachments.some((file) => file?.type && file?.type?.includes('image')) &&
-      visionModelAvailable &&
-      !this.isVisionModel
-    ) {
-      this.modelOptions.model = this.defaultVisionModel;
-      this.isVisionModel = true;
+    if (!attachments) {
+      return;
     }
 
+    const availableModels = this.options.modelsConfig?.[this.options.endpoint];
+    if (!availableModels) {
+      return;
+    }
+
+    let visionRequestDetected = false;
+    for (const file of attachments) {
+      if (file?.type?.includes('image')) {
+        visionRequestDetected = true;
+        break;
+      }
+    }
+    if (!visionRequestDetected) {
+      return;
+    }
+
+    this.isVisionModel = validateVisionModel({ model: this.modelOptions.model, availableModels });
     if (this.isVisionModel) {
       delete this.modelOptions.stop;
+      return;
     }
+
+    for (const model of availableModels) {
+      if (!validateVisionModel({ model, availableModels })) {
+        continue;
+      }
+      this.modelOptions.model = model;
+      this.isVisionModel = true;
+      delete this.modelOptions.stop;
+      return;
+    }
+
+    if (!availableModels.includes(this.defaultVisionModel)) {
+      return;
+    }
+    if (!validateVisionModel({ model: this.defaultVisionModel, availableModels })) {
+      return;
+    }
+
+    this.modelOptions.model = this.defaultVisionModel;
+    this.isVisionModel = true;
+    delete this.modelOptions.stop;
   }
 
   setupTokens() {
@@ -272,7 +309,7 @@ class OpenAIClient extends BaseClient {
     let tokenizer;
     this.encoding = 'text-davinci-003';
     if (this.isChatCompletion) {
-      this.encoding = 'cl100k_base';
+      this.encoding = this.modelOptions.model.includes('gpt-4o') ? 'o200k_base' : 'cl100k_base';
       tokenizer = this.constructor.getTokenizer(this.encoding);
     } else if (this.isUnofficialChatGptModel) {
       const extendSpecialTokens = {
@@ -377,6 +414,7 @@ class OpenAIClient extends BaseClient {
 
   getSaveOptions() {
     return {
+      maxContextTokens: this.options.maxContextTokens,
       chatGptLabel: this.options.chatGptLabel,
       promptPrefix: this.options.promptPrefix,
       resendFiles: this.options.resendFiles,
@@ -405,7 +443,11 @@ class OpenAIClient extends BaseClient {
    * @returns {Promise<MongoFile[]>}
    */
   async addImageURLs(message, attachments) {
-    const { files, image_urls } = await encodeAndFormat(this.options.req, attachments);
+    const { files, image_urls } = await encodeAndFormat(
+      this.options.req,
+      attachments,
+      this.options.endpoint,
+    );
     message.image_urls = image_urls.length ? image_urls : undefined;
     return files;
   }
@@ -547,12 +589,13 @@ class OpenAIClient extends BaseClient {
     let streamResult = null;
     this.modelOptions.user = this.user;
     const invalidBaseUrl = this.completionsUrl && extractBaseURL(this.completionsUrl) === null;
-    const useOldMethod = !!(invalidBaseUrl || !this.isChatCompletion || typeof Bun !== 'undefined');
+    const useOldMethod = !!(invalidBaseUrl || !this.isChatCompletion);
     if (typeof opts.onProgress === 'function' && useOldMethod) {
       const completionResult = await this.getCompletion(
         payload,
         (progressMessage) => {
           if (progressMessage === '[DONE]') {
+            updateTokenWebsocket('[DONE]');
             return;
           }
 
@@ -715,6 +758,12 @@ class OpenAIClient extends BaseClient {
    *                            In case of failure, it will return the default title, "New Chat".
    */
   async titleConvo({ text, conversationId, responseText = '' }) {
+    this.conversationId = conversationId;
+
+    if (this.options.attachments) {
+      delete this.options.attachments;
+    }
+
     let title = 'New Chat';
     const convo = `||>User:
 "${truncateText(text)}"
@@ -780,7 +829,7 @@ class OpenAIClient extends BaseClient {
 
       const instructionsPayload = [
         {
-          role: 'system',
+          role: this.options.titleMessageRole ?? 'system',
           content: `Please generate ${titleInstruction}
 
 ${convo}
@@ -793,13 +842,17 @@ ${convo}
 
       try {
         let useChatCompletion = true;
+
         if (this.options.reverseProxyUrl === CohereConstants.API_URL) {
           useChatCompletion = false;
         }
+
         title = (
           await this.sendPayload(instructionsPayload, { modelOptions, useChatCompletion })
         ).replaceAll('"', '');
+
         const completionTokens = this.getTokenCount(title);
+
         this.recordTokenUsage({ promptTokens, completionTokens, context: 'title' });
       } catch (e) {
         logger.error(
@@ -823,6 +876,7 @@ ${convo}
         context: 'title',
         tokenBuffer: 150,
       });
+
       title = await runTitleChain({ llm, text, convo, signal: this.abortController.signal });
     } catch (e) {
       if (e?.message?.toLowerCase()?.includes('abort')) {
@@ -960,9 +1014,9 @@ ${convo}
     await spendTokens(
       {
         context,
-        user: this.user,
         model: this.modelOptions.model,
         conversationId: this.conversationId,
+        user: this.user ?? this.options.req.user?.id,
         endpointTokenConfig: this.options.endpointTokenConfig,
       },
       { promptTokens, completionTokens },
@@ -1054,7 +1108,12 @@ ${convo}
       }
 
       if (this.azure || this.options.azure) {
-        // Azure does not accept `model` in the body, so we need to remove it.
+        /* Azure Bug, extremely short default `max_tokens` response */
+        if (!modelOptions.max_tokens && modelOptions.model === 'gpt-4-vision-preview') {
+          modelOptions.max_tokens = 4000;
+        }
+
+        /* Azure does not accept `model` in the body, so we need to remove it. */
         delete modelOptions.model;
 
         opts.baseURL = this.langchainProxy
@@ -1075,15 +1134,13 @@ ${convo}
       let chatCompletion;
       /** @type {OpenAI} */
       const openai = new OpenAI({
+        fetch: this.fetch,
         apiKey: this.apiKey,
         ...opts,
       });
 
-      /* 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) {
+      /* Re-orders system message to the top of the messages payload, as not allowed anywhere else */
+      if (modelOptions.messages && (opts.baseURL.includes('api.mistral.ai') || this.isOllama)) {
         const { messages } = modelOptions;
 
         const systemMessageIndex = messages.findIndex((msg) => msg.role === 'system');
@@ -1094,10 +1151,16 @@ ${convo}
         }
 
         modelOptions.messages = messages;
+      }
 
-        if (messages.length === 1 && messages[0].role === 'system') {
-          modelOptions.messages[0].role = 'user';
-        }
+      /* If there is only one message and it's a system message, change the role to user */
+      if (
+        (opts.baseURL.includes('api.mistral.ai') || opts.baseURL.includes('api.perplexity.ai')) &&
+        modelOptions.messages &&
+        modelOptions.messages.length === 1 &&
+        modelOptions.messages[0]?.role === 'system'
+      ) {
+        modelOptions.messages[0].role = 'user';
       }
 
       if (this.options.addParams && typeof this.options.addParams === 'object') {
@@ -1121,6 +1184,15 @@ ${convo}
         });
       }
 
+      if (this.message_file_map && this.isOllama) {
+        const ollamaClient = new OllamaClient({ baseURL });
+        return await ollamaClient.chatCompletion({
+          payload: modelOptions,
+          onProgress,
+          abortController,
+        });
+      }
+
       let UnexpectedRoleError = false;
       if (modelOptions.stream) {
         const stream = await openai.beta.chat.completions
@@ -1151,6 +1223,8 @@ ${convo}
             }
           });
 
+        const azureDelay = this.modelOptions.model?.includes('gpt-4') ? 30 : 17;
+
         for await (const chunk of stream) {
           const token = chunk.choices[0]?.delta?.content || '';
           intermediateReply += token;
@@ -1160,7 +1234,9 @@ ${convo}
             break;
           }
 
-          await sleep(25);
+          if (this.azure) {
+            await sleep(azureDelay);
+          }
         }
 
         if (!UnexpectedRoleError) {

api/app/clients/PluginsClient.js

@@ -250,6 +250,7 @@ class PluginsClient extends OpenAIClient {
       this.setOptions(opts);
       return super.sendMessage(message, opts);
     }
+
     logger.debug('[PluginsClient] sendMessage', { userMessageText: message, opts });
     const {
       user,
@@ -264,6 +265,14 @@ class PluginsClient extends OpenAIClient {
       onToolEnd,
     } = await this.handleStartMethods(message, opts);
 
+    if (opts.progressCallback) {
+      opts.onProgress = opts.progressCallback.call(null, {
+        ...(opts.progressOptions ?? {}),
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
+    }
+
     this.currentMessages.push(userMessage);
 
     let {

api/app/clients/prompts/createContextHandlers.js

@@ -8,8 +8,6 @@ In your response, remember to follow these guidelines:
 - If you don't know the answer, simply say that you don't know.
 - If you are unsure how to answer, ask for clarification.
 - Avoid mentioning that you obtained the information from the context.
-
-Answer appropriately in the user's language.
 `;
 
 function createContextHandlers(req, userMessageContent) {
@@ -94,37 +92,40 @@ function createContextHandlers(req, userMessageContent) {
 
       const resolvedQueries = await Promise.all(queryPromises);
 
-      const context = resolvedQueries
-        .map((queryResult, index) => {
-          const file = processedFiles[index];
-          let contextItems = queryResult.data;
+      const context =
+        resolvedQueries.length === 0
+          ? '\n\tThe semantic search did not return any results.'
+          : resolvedQueries
+            .map((queryResult, index) => {
+              const file = processedFiles[index];
+              let contextItems = queryResult.data;
 
-          const generateContext = (currentContext) =>
-            `
+              const generateContext = (currentContext) =>
+                `
           <file>
             <filename>${file.filename}</filename>
             <context>${currentContext}
             </context>
           </file>`;
 
-          if (useFullContext) {
-            return generateContext(`\n${contextItems}`);
-          }
+              if (useFullContext) {
+                return generateContext(`\n${contextItems}`);
+              }
 
-          contextItems = queryResult.data
-            .map((item) => {
-              const pageContent = item[0].page_content;
-              return `
+              contextItems = queryResult.data
+                .map((item) => {
+                  const pageContent = item[0].page_content;
+                  return `
             <contextItem>
               <![CDATA[${pageContent?.trim()}]]>
             </contextItem>`;
+                })
+                .join('');
+
+              return generateContext(contextItems);
             })
             .join('');
 
-          return generateContext(contextItems);
-        })
-        .join('');
-
       if (useFullContext) {
         const prompt = `${header}
           ${context}

api/app/clients/prompts/titlePrompts.js

@@ -28,7 +28,7 @@ ${convo}`,
 };
 
 const titleInstruction =
-  'a concise, 5-word-or-less title for the conversation, using its same language, with no punctuation. Apply title case conventions appropriate for the language. For English, use AP Stylebook Title Case. Never directly mention the language name or the word "title"';
+  'a concise, 5-word-or-less title for the conversation, using its same language, with no punctuation. Apply title case conventions appropriate for the language. Never directly mention the language name or the word "title"';
 const titleFunctionPrompt = `In this environment you have access to a set of tools you can use to generate the conversation title.
   
 You may call them like this:
@@ -59,25 +59,57 @@ Submit a brief title in the conversation's language, following the parameter des
 </tool_description>
 </tools>`;
 
+const genTranslationPrompt = (
+  translationPrompt,
+) => `In this environment you have access to a set of tools you can use to translate text.
+  
+You may call them like this:
+<function_calls>
+<invoke>
+<tool_name>$TOOL_NAME</tool_name>
+<parameters>
+<$PARAMETER_NAME>$PARAMETER_VALUE</$PARAMETER_NAME>
+...
+</parameters>
+</invoke>
+</function_calls>
+
+Here are the tools available:
+<tools>
+<tool_description>
+<tool_name>submit_translation</tool_name>
+<description>
+Submit a translation in the target language, following the parameter description and its language closely.
+</description>
+<parameters>
+<parameter>
+<name>translation</name>
+<type>string</type>
+<description>${translationPrompt}
+ONLY include the generated translation without quotations, nor its related key</description>
+</parameter>
+</parameters>
+</tool_description>
+</tools>`;
+
 /**
- * Parses titles from title functions based on the provided prompt.
- * @param {string} prompt - The prompt containing the title function.
- * @returns {string} The parsed title. "New Chat" if no title is found.
+ * Parses specified parameter from the provided prompt.
+ * @param {string} prompt - The prompt containing the desired parameter.
+ * @param {string} paramName - The name of the parameter to extract.
+ * @returns {string} The parsed parameter's value or a default value if not found.
  */
-function parseTitleFromPrompt(prompt) {
-  const titleRegex = /<title>(.+?)<\/title>/;
-  const titleMatch = prompt.match(titleRegex);
+function parseParamFromPrompt(prompt, paramName) {
+  const paramRegex = new RegExp(`<${paramName}>([\\s\\S]+?)</${paramName}>`);
+  const paramMatch = prompt.match(paramRegex);
 
-  if (titleMatch && titleMatch[1]) {
-    const title = titleMatch[1].trim();
-
-    // // Capitalize the first letter of each word; Note: unnecessary due to title case prompting
-    // const capitalizedTitle = title.replace(/\b\w/g, (char) => char.toUpperCase());
-
-    return title;
+  if (paramMatch && paramMatch[1]) {
+    return paramMatch[1].trim();
   }
 
-  return 'New Chat';
+  if (prompt && prompt.length) {
+    return `NO TOOL INVOCATION: ${prompt}`;
+  }
+  return `No ${paramName} provided`;
 }
 
 module.exports = {
@@ -85,5 +117,6 @@ module.exports = {
   titleInstruction,
   createTitlePrompt,
   titleFunctionPrompt,
-  parseTitleFromPrompt,
+  parseParamFromPrompt,
+  genTranslationPrompt,
 };

api/app/clients/specs/FakeClient.js

@@ -40,7 +40,8 @@ class FakeClient extends BaseClient {
       };
     }
 
-    this.maxContextTokens = getModelMaxTokens(this.modelOptions.model) ?? 4097;
+    this.maxContextTokens =
+      this.options.maxContextTokens ?? getModelMaxTokens(this.modelOptions.model) ?? 4097;
   }
   buildMessages() {}
   getTokenCount(str) {

api/app/clients/specs/OpenAIClient.test.js

@@ -144,6 +144,7 @@ describe('OpenAIClient', () => {
 
   const defaultOptions = {
     // debug: true,
+    req: {},
     openaiApiKey: 'new-api-key',
     modelOptions: {
       model,
@@ -157,12 +158,19 @@ describe('OpenAIClient', () => {
     azureOpenAIApiVersion: '2020-07-01-preview',
   };
 
+  let originalWarn;
+
   beforeAll(() => {
-    jest.spyOn(console, 'warn').mockImplementation(() => {});
+    originalWarn = console.warn;
+    console.warn = jest.fn();
   });
 
   afterAll(() => {
-    console.warn.mockRestore();
+    console.warn = originalWarn;
+  });
+
+  beforeEach(() => {
+    console.warn.mockClear();
   });
 
   beforeEach(() => {
@@ -662,4 +670,35 @@ describe('OpenAIClient', () => {
       expect(constructorArgs.baseURL).toBe(expectedURL);
     });
   });
+
+  describe('checkVisionRequest functionality', () => {
+    let client;
+    const attachments = [{ type: 'image/png' }];
+
+    beforeEach(() => {
+      client = new OpenAIClient('test-api-key', {
+        endpoint: 'ollama',
+        modelOptions: {
+          model: 'initial-model',
+        },
+        modelsConfig: {
+          ollama: ['initial-model', 'llava', 'other-model'],
+        },
+      });
+
+      client.defaultVisionModel = 'non-valid-default-model';
+    });
+
+    afterEach(() => {
+      jest.restoreAllMocks();
+    });
+
+    it('should set "llava" as the model if it is the first valid model when default validation fails', () => {
+      client.checkVisionRequest(attachments);
+
+      expect(client.modelOptions.model).toBe('llava');
+      expect(client.isVisionModel).toBeTruthy();
+      expect(client.modelOptions.stop).toBeUndefined();
+    });
+  });
 });

api/app/clients/tools/structured/StableDiffusion.js

@@ -80,13 +80,18 @@ class StableDiffusionAPI extends StructuredTool {
     const payload = {
       prompt,
       negative_prompt,
-      sampler_index: 'DPM++ 2M Karras',
       cfg_scale: 4.5,
       steps: 22,
       width: 1024,
       height: 1024,
     };
-    const generationResponse = await axios.post(`${url}/sdapi/v1/txt2img`, payload);
+    let generationResponse;
+    try {
+      generationResponse = await axios.post(`${url}/sdapi/v1/txt2img`, payload);
+    } catch (error) {
+      logger.error('[StableDiffusion] Error while generating image:', error);
+      return 'Error making API request.';
+    }
     const image = generationResponse.data.images[0];
 
     /** @type {{ height: number, width: number, seed: number, infotexts: string[] }} */

api/cache/getLogStores.js

@@ -7,6 +7,7 @@ const keyvMongo = require('./keyvMongo');
 
 const { BAN_DURATION, USE_REDIS } = process.env ?? {};
 const THIRTY_MINUTES = 1800000;
+const TEN_MINUTES = 600000;
 
 const duration = math(BAN_DURATION, 7200000);
 
@@ -24,6 +25,10 @@ const config = isEnabled(USE_REDIS)
   ? new Keyv({ store: keyvRedis })
   : new Keyv({ namespace: CacheKeys.CONFIG_STORE });
 
+const audioRuns = isEnabled(USE_REDIS) // ttl: 30 minutes
+  ? new Keyv({ store: keyvRedis, ttl: TEN_MINUTES })
+  : new Keyv({ namespace: CacheKeys.AUDIO_RUNS, ttl: TEN_MINUTES });
+
 const tokenConfig = isEnabled(USE_REDIS) // ttl: 30 minutes
   ? new Keyv({ store: keyvRedis, ttl: THIRTY_MINUTES })
   : new Keyv({ namespace: CacheKeys.TOKEN_CONFIG, ttl: THIRTY_MINUTES });
@@ -55,7 +60,13 @@ const namespaces = {
   message_limit: createViolationInstance('message_limit'),
   token_balance: createViolationInstance(ViolationTypes.TOKEN_BALANCE),
   registrations: createViolationInstance('registrations'),
+  [ViolationTypes.TTS_LIMIT]: createViolationInstance(ViolationTypes.TTS_LIMIT),
+  [ViolationTypes.STT_LIMIT]: createViolationInstance(ViolationTypes.STT_LIMIT),
   [ViolationTypes.FILE_UPLOAD_LIMIT]: createViolationInstance(ViolationTypes.FILE_UPLOAD_LIMIT),
+  [ViolationTypes.VERIFY_EMAIL_LIMIT]: createViolationInstance(ViolationTypes.VERIFY_EMAIL_LIMIT),
+  [ViolationTypes.RESET_PASSWORD_LIMIT]: createViolationInstance(
+    ViolationTypes.RESET_PASSWORD_LIMIT,
+  ),
   [ViolationTypes.ILLEGAL_MODEL_REQUEST]: createViolationInstance(
     ViolationTypes.ILLEGAL_MODEL_REQUEST,
   ),
@@ -64,6 +75,7 @@ const namespaces = {
   [CacheKeys.TOKEN_CONFIG]: tokenConfig,
   [CacheKeys.GEN_TITLE]: genTitle,
   [CacheKeys.MODEL_QUERIES]: modelQueries,
+  [CacheKeys.AUDIO_RUNS]: audioRuns,
 };
 
 /**

api/cache/logViolation.js

@@ -1,6 +1,6 @@
+const { isEnabled } = require('~/server/utils');
 const getLogStores = require('./getLogStores');
 const banViolation = require('./banViolation');
-const { isEnabled } = require('../server/utils');
 
 /**
  * Logs the violation.

api/config/parsers.js

@@ -27,26 +27,25 @@ function getMatchingSensitivePatterns(valueStr) {
 }
 
 /**
- * Redacts sensitive information from a console message.
- *
+ * Redacts sensitive information from a console message and trims it to a specified length if provided.
  * @param {string} str - The console message to be redacted.
- * @returns {string} - The redacted console message.
+ * @param {number} [trimLength] - The optional length at which to trim the redacted message.
+ * @returns {string} - The redacted and optionally trimmed console message.
  */
-function redactMessage(str) {
+function redactMessage(str, trimLength) {
   if (!str) {
     return '';
   }
 
   const patterns = getMatchingSensitivePatterns(str);
-
-  if (patterns.length === 0) {
-    return str;
-  }
-
   patterns.forEach((pattern) => {
     str = str.replace(pattern, '$1[REDACTED]');
   });
 
+  if (trimLength !== undefined && str.length > trimLength) {
+    return `${str.substring(0, trimLength)}...`;
+  }
+
   return str;
 }
 

api/models/Action.js

@@ -62,8 +62,24 @@ const deleteAction = async (searchParams, session = null) => {
   return await Action.findOneAndDelete(searchParams, options).lean();
 };
 
-module.exports = {
-  updateAction,
-  getActions,
-  deleteAction,
+/**
+ * Deletes actions by params, within a transaction session if provided.
+ *
+ * @param {Object} searchParams - The search parameters to find the actions to delete.
+ * @param {string} searchParams.action_id - The ID of the action(s) to delete.
+ * @param {string} searchParams.user - The user ID of the action's author.
+ * @param {mongoose.ClientSession} [session] - The transaction session to use (optional).
+ * @returns {Promise<Number>} A promise that resolves to the number of deleted action documents.
+ */
+const deleteActions = async (searchParams, session = null) => {
+  const options = session ? { session } : {};
+  const result = await Action.deleteMany(searchParams, options);
+  return result.deletedCount;
+};
+
+module.exports = {
+  getActions,
+  updateAction,
+  deleteAction,
+  deleteActions,
 };

api/models/Assistant.js

@@ -14,7 +14,7 @@ const Assistant = mongoose.model('assistant', assistantSchema);
  * @param {mongoose.ClientSession} [session] - The transaction session to use (optional).
  * @returns {Promise<Object>} The updated or newly created assistant document as a plain object.
  */
-const updateAssistant = async (searchParams, updateData, session = null) => {
+const updateAssistantDoc = async (searchParams, updateData, session = null) => {
   const options = { new: true, upsert: true, session };
   return await Assistant.findOneAndUpdate(searchParams, updateData, options).lean();
 };
@@ -39,8 +39,21 @@ const getAssistants = async (searchParams) => {
   return await Assistant.find(searchParams).lean();
 };
 
+/**
+ * Deletes an assistant based on the provided ID.
+ *
+ * @param {Object} searchParams - The search parameters to find the assistant to delete.
+ * @param {string} searchParams.assistant_id - The ID of the assistant to delete.
+ * @param {string} searchParams.user - The user ID of the assistant's author.
+ * @returns {Promise<void>} Resolves when the assistant has been successfully deleted.
+ */
+const deleteAssistant = async (searchParams) => {
+  return await Assistant.findOneAndDelete(searchParams);
+};
+
 module.exports = {
-  updateAssistant,
+  updateAssistantDoc,
+  deleteAssistant,
   getAssistants,
   getAssistant,
 };

api/models/Conversation.js

@@ -21,7 +21,7 @@ module.exports = {
   Conversation,
   saveConvo: async (user, { conversationId, newConversationId, ...convo }) => {
     try {
-      const messages = await getMessages({ conversationId });
+      const messages = await getMessages({ conversationId }, '_id');
       const update = { ...convo, messages, user };
       if (newConversationId) {
         update.conversationId = newConversationId;

api/models/File.js

@@ -97,8 +97,12 @@ const deleteFileByFilter = async (filter) => {
  * @param {Array<string>} file_ids - The unique identifiers of the files to delete.
  * @returns {Promise<Object>} A promise that resolves to the result of the deletion operation.
  */
-const deleteFiles = async (file_ids) => {
-  return await File.deleteMany({ file_id: { $in: file_ids } });
+const deleteFiles = async (file_ids, user) => {
+  let deleteQuery = { file_id: { $in: file_ids } };
+  if (user) {
+    deleteQuery = { user: user };
+  }
+  return await File.deleteMany(deleteQuery);
 };
 
 module.exports = {

api/models/Message.js

@@ -129,6 +129,14 @@ module.exports = {
       throw new Error('Failed to save message.');
     }
   },
+  async updateMessageText({ messageId, text }) {
+    try {
+      await Message.updateOne({ messageId }, { text });
+    } catch (err) {
+      logger.error('Error updating message text:', err);
+      throw new Error('Failed to update message text.');
+    }
+  },
   async updateMessage(message) {
     try {
       const { messageId, ...update } = message;
@@ -171,8 +179,18 @@ module.exports = {
     }
   },
 
-  async getMessages(filter) {
+  /**
+   * Retrieves messages from the database.
+   * @param {Record<string, unknown>} filter
+   * @param {string | undefined} [select]
+   * @returns
+   */
+  async getMessages(filter, select) {
     try {
+      if (select) {
+        return await Message.find(filter).select(select).sort({ createdAt: 1 }).lean();
+      }
+
       return await Message.find(filter).sort({ createdAt: 1 }).lean();
     } catch (err) {
       logger.error('Error getting messages:', err);

api/models/Share.js

@@ -0,0 +1,106 @@
+const crypto = require('crypto');
+const { getMessages } = require('./Message');
+const SharedLink = require('./schema/shareSchema');
+const logger = require('~/config/winston');
+
+module.exports = {
+  SharedLink,
+  getSharedMessages: async (shareId) => {
+    try {
+      const share = await SharedLink.findOne({ shareId })
+        .populate({
+          path: 'messages',
+          select: '-_id -__v -user',
+        })
+        .select('-_id -__v -user')
+        .lean();
+
+      if (!share || !share.conversationId || !share.isPublic) {
+        return null;
+      }
+
+      return share;
+    } catch (error) {
+      logger.error('[getShare] Error getting share link', error);
+      return { message: 'Error getting share link' };
+    }
+  },
+
+  getSharedLinks: async (user, pageNumber = 1, pageSize = 25, isPublic = true) => {
+    const query = { user, isPublic };
+    try {
+      const totalConvos = (await SharedLink.countDocuments(query)) || 1;
+      const totalPages = Math.ceil(totalConvos / pageSize);
+      const shares = await SharedLink.find(query)
+        .sort({ updatedAt: -1 })
+        .skip((pageNumber - 1) * pageSize)
+        .limit(pageSize)
+        .select('-_id -__v -user')
+        .lean();
+
+      return { sharedLinks: shares, pages: totalPages, pageNumber, pageSize };
+    } catch (error) {
+      logger.error('[getShareByPage] Error getting shares', error);
+      return { message: 'Error getting shares' };
+    }
+  },
+
+  createSharedLink: async (user, { conversationId, ...shareData }) => {
+    const share = await SharedLink.findOne({ conversationId }).select('-_id -__v -user').lean();
+    if (share) {
+      return share;
+    }
+
+    try {
+      const shareId = crypto.randomUUID();
+      const messages = await getMessages({ conversationId });
+      const update = { ...shareData, shareId, messages, user };
+      return await SharedLink.findOneAndUpdate({ conversationId: conversationId, user }, update, {
+        new: true,
+        upsert: true,
+      });
+    } catch (error) {
+      logger.error('[saveShareMessage] Error saving conversation', error);
+      return { message: 'Error saving conversation' };
+    }
+  },
+
+  updateSharedLink: async (user, { conversationId, ...shareData }) => {
+    const share = await SharedLink.findOne({ conversationId }).select('-_id -__v -user').lean();
+    if (!share) {
+      return { message: 'Share not found' };
+    }
+    // update messages to the latest
+    const messages = await getMessages({ conversationId });
+    const update = { ...shareData, messages, user };
+    return await SharedLink.findOneAndUpdate({ conversationId: conversationId, user }, update, {
+      new: true,
+      upsert: false,
+    });
+  },
+
+  deleteSharedLink: async (user, { shareId }) => {
+    const share = await SharedLink.findOne({ shareId, user });
+    if (!share) {
+      return { message: 'Share not found' };
+    }
+    return await SharedLink.findOneAndDelete({ shareId, user });
+  },
+  /**
+   * Deletes all shared links for a specific user.
+   * @param {string} user - The user ID.
+   * @returns {Promise<{ message: string, deletedCount?: number }>} A result object indicating success or error message.
+   */
+  deleteAllSharedLinks: async (user) => {
+    try {
+      const result = await SharedLink.deleteMany({ user });
+      return {
+        message: 'All shared links have been deleted successfully',
+        deletedCount: result.deletedCount,
+      };
+    } catch (error) {
+      logger.error('[deleteAllSharedLinks] Error deleting shared links', error);
+      return { message: 'Error deleting shared links' };
+    }
+  },
+};

api/models/User.js

@@ -1,61 +1,5 @@
 const mongoose = require('mongoose');
-const bcrypt = require('bcryptjs');
-const signPayload = require('../server/services/signPayload');
-const userSchema = require('./schema/userSchema.js');
-const { SESSION_EXPIRY } = process.env ?? {};
-const expires = eval(SESSION_EXPIRY) ?? 1000 * 60 * 15;
-
-userSchema.methods.toJSON = function () {
-  return {
-    id: this._id,
-    provider: this.provider,
-    email: this.email,
-    name: this.name,
-    username: this.username,
-    avatar: this.avatar,
-    role: this.role,
-    emailVerified: this.emailVerified,
-    plugins: this.plugins,
-    createdAt: this.createdAt,
-    updatedAt: this.updatedAt,
-  };
-};
-
-userSchema.methods.generateToken = async function () {
-  return await signPayload({
-    payload: {
-      id: this._id,
-      username: this.username,
-      provider: this.provider,
-      email: this.email,
-    },
-    secret: process.env.JWT_SECRET,
-    expirationTime: expires / 1000,
-  });
-};
-
-userSchema.methods.comparePassword = function (candidatePassword, callback) {
-  bcrypt.compare(candidatePassword, this.password, (err, isMatch) => {
-    if (err) {
-      return callback(err);
-    }
-    callback(null, isMatch);
-  });
-};
-
-module.exports.hashPassword = async (password) => {
-  const hashedPassword = await new Promise((resolve, reject) => {
-    bcrypt.hash(password, 10, function (err, hash) {
-      if (err) {
-        reject(err);
-      } else {
-        resolve(hash);
-      }
-    });
-  });
-
-  return hashedPassword;
-};
+const userSchema = require('~/models/schema/userSchema');
 
 const User = mongoose.model('User', userSchema);
 

api/models/index.js

@@ -6,9 +6,18 @@ const {
   deleteMessagesSince,
   deleteMessages,
 } = require('./Message');
+const {
+  comparePassword,
+  deleteUserById,
+  generateToken,
+  getUserById,
+  updateUser,
+  createUser,
+  countUsers,
+  findUser,
+} = require('./userMethods');
 const { getConvoTitle, getConvo, saveConvo, deleteConvos } = require('./Conversation');
 const { getPreset, getPresets, savePreset, deletePresets } = require('./Preset');
-const { hashPassword, getUser, updateUser } = require('./userMethods');
 const {
   findFileById,
   createFile,
@@ -29,9 +38,14 @@ module.exports = {
   Session,
   Balance,
 
-  hashPassword,
+  comparePassword,
+  deleteUserById,
+  generateToken,
+  getUserById,
+  countUsers,
+  createUser,
   updateUser,
-  getUser,
+  findUser,
 
   getMessages,
   saveMessage,

api/models/plugins/mongoMeili.js

@@ -155,7 +155,7 @@ const createMeiliMongooseModel = function ({ index, attributesToIndex }) {
             function (results, value, key) {
               return { ...results, [key]: 1 };
             },
-            { _id: 1 },
+            { _id: 1, __v: 1 },
           ),
         ).lean();
 
@@ -348,7 +348,7 @@ module.exports = function mongoMeili(schema, options) {
       try {
         meiliDoc = await client.index('convos').getDocument(doc.conversationId);
       } catch (error) {
-        logger.error(
+        logger.debug(
           '[MeiliMongooseModel.findOneAndUpdate] Convo not found in MeiliSearch and will index ' +
             doc.conversationId,
           error,

api/models/schema/defaults.js

@@ -104,6 +104,12 @@ const conversationPreset = {
     type: String,
   },
   tools: { type: [{ type: String }], default: undefined },
+  maxContextTokens: {
+    type: Number,
+  },
+  max_tokens: {
+    type: Number,
+  },
 };
 
 const agentOptions = {

api/models/schema/fileSchema.js

@@ -3,9 +3,9 @@ const mongoose = require('mongoose');
 
 /**
  * @typedef {Object} MongoFile
- * @property {mongoose.Schema.Types.ObjectId} [_id] - MongoDB Document ID
+ * @property {ObjectId} [_id] - MongoDB Document ID
  * @property {number} [__v] - MongoDB Version Key
- * @property {mongoose.Schema.Types.ObjectId} user - User ID
+ * @property {ObjectId} user - User ID
  * @property {string} [conversationId] - Optional conversation ID
  * @property {string} file_id - File identifier
  * @property {string} [temp_file_id] - Temporary File identifier
@@ -14,17 +14,19 @@ const mongoose = require('mongoose');
  * @property {string} filepath - Location of the file
  * @property {'file'} object - Type of object, always 'file'
  * @property {string} type - Type of file
- * @property {number} usage - Number of uses of the file
+ * @property {number} [usage=0] - Number of uses of the file
  * @property {string} [context] - Context of the file origin
- * @property {boolean} [embedded] - Whether or not the file is embedded in vector db
+ * @property {boolean} [embedded=false] - Whether or not the file is embedded in vector db
  * @property {string} [model] - The model to identify the group region of the file (for Azure OpenAI hosting)
- * @property {string} [source] - The source of the file
+ * @property {string} [source] - The source of the file (e.g., from FileSources)
  * @property {number} [width] - Optional width of the file
  * @property {number} [height] - Optional height of the file
- * @property {Date} [expiresAt] - Optional height of the file
+ * @property {Date} [expiresAt] - Optional expiration date of the file
  * @property {Date} [createdAt] - Date when the file was created
  * @property {Date} [updatedAt] - Date when the file was updated
  */
+
+/** @type {MongooseSchema<MongoFile>} */
 const fileSchema = mongoose.Schema(
   {
     user: {
@@ -91,7 +93,7 @@ const fileSchema = mongoose.Schema(
     height: Number,
     expiresAt: {
       type: Date,
-      expires: 3600,
+      expires: 3600, // 1 hour in seconds
     },
   },
   {

api/models/schema/messageSchema.js

@@ -11,6 +11,7 @@ const messageSchema = mongoose.Schema(
     },
     conversationId: {
       type: String,
+      index: true,
       required: true,
       meiliIndex: true,
     },

api/models/schema/shareSchema.js

@@ -0,0 +1,38 @@
+const mongoose = require('mongoose');
+
+const shareSchema = mongoose.Schema(
+  {
+    conversationId: {
+      type: String,
+      required: true,
+    },
+    title: {
+      type: String,
+      index: true,
+    },
+    user: {
+      type: String,
+      index: true,
+    },
+    messages: [{ type: mongoose.Schema.Types.ObjectId, ref: 'Message' }],
+    shareId: {
+      type: String,
+      index: true,
+    },
+    isPublic: {
+      type: Boolean,
+      default: false,
+    },
+    isVisible: {
+      type: Boolean,
+      default: false,
+    },
+    isAnonymous: {
+      type: Boolean,
+      default: true,
+    },
+  },
+  { timestamps: true },
+);
+
+module.exports = mongoose.model('SharedLink', shareSchema);

api/models/schema/tokenSchema.js

@@ -7,6 +7,9 @@ const tokenSchema = new Schema({
     required: true,
     ref: 'user',
   },
+  email: {
+    type: String,
+  },
   token: {
     type: String,
     required: true,

api/models/schema/userSchema.js

@@ -1,5 +1,35 @@
 const mongoose = require('mongoose');
 
+/**
+ * @typedef {Object} MongoSession
+ * @property {string} [refreshToken] - The refresh token
+ */
+
+/**
+ * @typedef {Object} MongoUser
+ * @property {ObjectId} [_id] - MongoDB Document ID
+ * @property {string} [name] - The user's name
+ * @property {string} [username] - The user's username, in lowercase
+ * @property {string} email - The user's email address
+ * @property {boolean} emailVerified - Whether the user's email is verified
+ * @property {string} [password] - The user's password, trimmed with 8-128 characters
+ * @property {string} [avatar] - The URL of the user's avatar
+ * @property {string} provider - The provider of the user's account (e.g., 'local', 'google')
+ * @property {string} [role='USER'] - The role of the user
+ * @property {string} [googleId] - Optional Google ID for the user
+ * @property {string} [facebookId] - Optional Facebook ID for the user
+ * @property {string} [openidId] - Optional OpenID ID for the user
+ * @property {string} [ldapId] - Optional LDAP ID for the user
+ * @property {string} [githubId] - Optional GitHub ID for the user
+ * @property {string} [discordId] - Optional Discord ID for the user
+ * @property {Array} [plugins=[]] - List of plugins used by the user
+ * @property {Array.<MongoSession>} [refreshToken] - List of sessions with refresh tokens
+ * @property {Date} [expiresAt] - Optional expiration date of the file
+ * @property {Date} [createdAt] - Date when the user was created (added by timestamps)
+ * @property {Date} [updatedAt] - Date when the user was last updated (added by timestamps)
+ */
+
+/** @type {MongooseSchema<MongoSession>} */
 const Session = mongoose.Schema({
   refreshToken: {
     type: String,
@@ -7,6 +37,7 @@ const Session = mongoose.Schema({
   },
 });
 
+/** @type {MongooseSchema<MongoUser>} */
 const userSchema = mongoose.Schema(
   {
     name: {
@@ -64,6 +95,11 @@ const userSchema = mongoose.Schema(
       unique: true,
       sparse: true,
     },
+    ldapId: {
+      type: String,
+      unique: true,
+      sparse: true,
+    },
     githubId: {
       type: String,
       unique: true,
@@ -81,6 +117,10 @@ const userSchema = mongoose.Schema(
     refreshToken: {
       type: [Session],
     },
+    expiresAt: {
+      type: Date,
+      expires: 604800, // 7 days in seconds
+    },
   },
   { timestamps: true },
 );

api/models/spendTokens.js

@@ -40,7 +40,7 @@ const spendTokens = async (txData, tokenUsage) => {
       });
     }
 
-    if (!completionTokens) {
+    if (!completionTokens && isNaN(completionTokens)) {
       logger.debug('[spendTokens] !completionTokens', { prompt, completion });
       return;
     }

api/models/tx.js

@@ -12,6 +12,7 @@ const tokenValues = {
   '4k': { prompt: 1.5, completion: 2 },
   '16k': { prompt: 3, completion: 4 },
   'gpt-3.5-turbo-1106': { prompt: 1, completion: 2 },
+  'gpt-4o': { prompt: 5, completion: 15 },
   'gpt-4-1106': { prompt: 10, completion: 30 },
   'gpt-3.5-turbo-0125': { prompt: 0.5, completion: 1.5 },
   'claude-3-opus': { prompt: 15, completion: 75 },
@@ -52,6 +53,8 @@ const getValueKey = (model, endpoint) => {
     return 'gpt-3.5-turbo-1106';
   } else if (modelName.includes('gpt-3.5')) {
     return '4k';
+  } else if (modelName.includes('gpt-4o')) {
+    return 'gpt-4o';
   } else if (modelName.includes('gpt-4-vision')) {
     return 'gpt-4-1106';
   } else if (modelName.includes('gpt-4-1106')) {

api/models/tx.spec.js

@@ -41,6 +41,13 @@ describe('getValueKey', () => {
     expect(getValueKey('gpt-4-turbo')).toBe('gpt-4-1106');
     expect(getValueKey('gpt-4-0125')).toBe('gpt-4-1106');
   });
+
+  it('should return "gpt-4o" for model type of "gpt-4o"', () => {
+    expect(getValueKey('gpt-4o-2024-05-13')).toBe('gpt-4o');
+    expect(getValueKey('openai/gpt-4o')).toBe('gpt-4o');
+    expect(getValueKey('gpt-4o-turbo')).toBe('gpt-4o');
+    expect(getValueKey('gpt-4o-0125')).toBe('gpt-4o');
+  });
 });
 
 describe('getMultiplier', () => {
@@ -84,6 +91,17 @@ describe('getMultiplier', () => {
     );
   });
 
+  it('should return the correct multiplier for gpt-4o', () => {
+    const valueKey = getValueKey('gpt-4o-2024-05-13');
+    expect(getMultiplier({ valueKey, tokenType: 'prompt' })).toBe(tokenValues['gpt-4o'].prompt);
+    expect(getMultiplier({ valueKey, tokenType: 'completion' })).toBe(
+      tokenValues['gpt-4o'].completion,
+    );
+    expect(getMultiplier({ valueKey, tokenType: 'completion' })).not.toBe(
+      tokenValues['gpt-4-1106'].completion,
+    );
+  });
+
   it('should derive the valueKey from the model if not provided for new models', () => {
     expect(
       getMultiplier({ tokenType: 'prompt', model: 'gpt-3.5-turbo-1106-some-other-info' }),

api/models/userMethods.js

@@ -1,28 +1,37 @@
 const bcrypt = require('bcryptjs');
+const signPayload = require('~/server/services/signPayload');
 const User = require('./User');
 
-const hashPassword = async (password) => {
-  const hashedPassword = await new Promise((resolve, reject) => {
-    bcrypt.hash(password, 10, function (err, hash) {
-      if (err) {
-        reject(err);
-      } else {
-        resolve(hash);
-      }
-    });
-  });
-
-  return hashedPassword;
-};
-
 /**
  * Retrieve a user by ID and convert the found user document to a plain object.
  *
  * @param {string} userId - The ID of the user to find and return as a plain object.
- * @returns {Promise<Object>} A plain object representing the user document, or `null` if no user is found.
+ * @param {string|string[]} [fieldsToSelect] - The fields to include or exclude in the returned document.
+ * @returns {Promise<MongoUser>} A plain object representing the user document, or `null` if no user is found.
  */
-const getUser = async function (userId) {
-  return await User.findById(userId).lean();
+const getUserById = async function (userId, fieldsToSelect = null) {
+  const query = User.findById(userId);
+
+  if (fieldsToSelect) {
+    query.select(fieldsToSelect);
+  }
+
+  return await query.lean();
+};
+
+/**
+ * Search for a single user based on partial data and return matching user document as plain object.
+ * @param {Partial<MongoUser>} searchCriteria - The partial data to use for searching the user.
+ * @param {string|string[]} [fieldsToSelect] - The fields to include or exclude in the returned document.
+ * @returns {Promise<MongoUser>} A plain object representing the user document, or `null` if no user is found.
+ */
+const findUser = async function (searchCriteria, fieldsToSelect = null) {
+  const query = User.findOne(searchCriteria);
+  if (fieldsToSelect) {
+    query.select(fieldsToSelect);
+  }
+
+  return await query.lean();
 };
 
 /**
@@ -30,17 +39,132 @@ const getUser = async function (userId) {
  *
  * @param {string} userId - The ID of the user to update.
  * @param {Object} updateData - An object containing the properties to update.
- * @returns {Promise<Object>} The updated user document as a plain object, or `null` if no user is found.
+ * @returns {Promise<MongoUser>} The updated user document as a plain object, or `null` if no user is found.
  */
 const updateUser = async function (userId, updateData) {
-  return await User.findByIdAndUpdate(userId, updateData, {
+  const updateOperation = {
+    $set: updateData,
+    $unset: { expiresAt: '' }, // Remove the expiresAt field to prevent TTL
+  };
+  return await User.findByIdAndUpdate(userId, updateOperation, {
     new: true,
     runValidators: true,
   }).lean();
 };
 
-module.exports = {
-  hashPassword,
-  updateUser,
-  getUser,
+/**
+ * Creates a new user, optionally with a TTL of 1 week.
+ * @param {MongoUser} data - The user data to be created, must contain user_id.
+ * @param {boolean} [disableTTL=true] - Whether to disable the TTL. Defaults to `true`.
+ * @returns {Promise<ObjectId>} A promise that resolves to the created user document ID.
+ * @throws {Error} If a user with the same user_id already exists.
+ */
+const createUser = async (data, disableTTL = true) => {
+  const userData = {
+    ...data,
+    expiresAt: disableTTL ? null : new Date(Date.now() + 604800 * 1000), // 1 week in milliseconds
+  };
+
+  if (disableTTL) {
+    delete userData.expiresAt;
+  }
+
+  try {
+    const user = await User.create(userData);
+    return user._id;
+  } catch (error) {
+    if (error.code === 11000) {
+      // Duplicate key error code
+      throw new Error(`User with \`_id\` ${data._id} already exists.`);
+    } else {
+      throw error;
+    }
+  }
+};
+
+/**
+ * Count the number of user documents in the collection based on the provided filter.
+ *
+ * @param {Object} [filter={}] - The filter to apply when counting the documents.
+ * @returns {Promise<number>} The count of documents that match the filter.
+ */
+const countUsers = async function (filter = {}) {
+  return await User.countDocuments(filter);
+};
+
+/**
+ * Delete a user by their unique ID.
+ *
+ * @param {string} userId - The ID of the user to delete.
+ * @returns {Promise<{ deletedCount: number }>} An object indicating the number of deleted documents.
+ */
+const deleteUserById = async function (userId) {
+  try {
+    const result = await User.deleteOne({ _id: userId });
+    if (result.deletedCount === 0) {
+      return { deletedCount: 0, message: 'No user found with that ID.' };
+    }
+    return { deletedCount: result.deletedCount, message: 'User was deleted successfully.' };
+  } catch (error) {
+    throw new Error('Error deleting user: ' + error.message);
+  }
+};
+
+const { SESSION_EXPIRY } = process.env ?? {};
+const expires = eval(SESSION_EXPIRY) ?? 1000 * 60 * 15;
+
+/**
+ * Generates a JWT token for a given user.
+ *
+ * @param {MongoUser} user - ID of the user for whom the token is being generated.
+ * @returns {Promise<string>} A promise that resolves to a JWT token.
+ */
+const generateToken = async (user) => {
+  if (!user) {
+    throw new Error('No user provided');
+  }
+
+  return await signPayload({
+    payload: {
+      id: user._id,
+      username: user.username,
+      provider: user.provider,
+      email: user.email,
+    },
+    secret: process.env.JWT_SECRET,
+    expirationTime: expires / 1000,
+  });
+};
+
+/**
+ * Compares the provided password with the user's password.
+ *
+ * @param {MongoUser} user - the user to compare password for.
+ * @param {string} candidatePassword - The password to test against the user's password.
+ * @returns {Promise<boolean>} A promise that resolves to a boolean indicating if the password matches.
+ */
+const comparePassword = async (user, candidatePassword) => {
+  if (!user) {
+    throw new Error('No user provided');
+  }
+
+  return new Promise((resolve, reject) => {
+    bcrypt.compare(candidatePassword, user.password, (err, isMatch) => {
+      if (err) {
+        reject(err);
+      }
+      resolve(isMatch);
+    });
+  });
+};
+
+module.exports = {
+  comparePassword,
+  deleteUserById,
+  generateToken,
+  getUserById,
+  countUsers,
+  createUser,
+  updateUser,
+  findUser,
 };

api/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@librechat/backend",
-  "version": "0.7.1",
+  "version": "0.7.3",
   "description": "",
   "scripts": {
     "start": "echo 'please run this from the root directory'",
@@ -40,8 +40,7 @@
     "@keyv/redis": "^2.8.1",
     "@langchain/community": "^0.0.46",
     "@langchain/google-genai": "^0.0.11",
-    "@langchain/google-vertexai": "^0.0.5",
-    "agenda": "^5.0.0",
+    "@langchain/google-vertexai": "^0.0.17",
     "axios": "^1.3.4",
     "bcryptjs": "^2.4.3",
     "cheerio": "^1.0.0-rc.12",
@@ -75,7 +74,8 @@
     "multer": "^1.4.5-lts.1",
     "nodejs-gpt": "^1.37.4",
     "nodemailer": "^6.9.4",
-    "openai": "4.36.0",
+    "ollama": "^0.5.0",
+    "openai": "^4.47.1",
     "openai-chat-tokens": "^0.2.8",
     "openid-client": "^5.4.2",
     "passport": "^0.6.0",
@@ -85,14 +85,16 @@
     "passport-github2": "^0.1.12",
     "passport-google-oauth20": "^2.0.0",
     "passport-jwt": "^4.0.1",
+    "passport-ldapauth": "^3.0.1",
     "passport-local": "^1.0.0",
     "pino": "^8.12.1",
     "sharp": "^0.32.6",
-    "tiktoken": "^1.0.10",
+    "tiktoken": "^1.0.15",
     "traverse": "^0.6.7",
     "ua-parser-js": "^1.0.36",
     "winston": "^3.11.0",
     "winston-daily-rotate-file": "^4.7.1",
+    "ws": "^8.17.0",
     "zod": "^3.22.4"
   },
   "devDependencies": {

api/server/controllers/AskController.js

@@ -105,11 +105,12 @@ const AskController = async (req, res, next, initializeClient, addTitle) => {
       getReqData,
       onStart,
       abortController,
-      onProgress: progressCallback.call(null, {
+      progressCallback,
+      progressOptions: {
         res,
         text,
-        parentMessageId: overrideParentMessageId || userMessageId,
-      }),
+        // parentMessageId: overrideParentMessageId || userMessageId,
+      },
     };
 
     let response = await client.sendMessage(text, messageOptions);

api/server/controllers/AuthController.js

@@ -1,45 +1,29 @@
 const crypto = require('crypto');
 const cookies = require('cookie');
 const jwt = require('jsonwebtoken');
-const { Session, User } = require('~/models');
 const {
   registerUser,
   resetPassword,
   setAuthTokens,
   requestPasswordReset,
 } = require('~/server/services/AuthService');
+const { Session, getUserById } = require('~/models');
 const { logger } = require('~/config');
 
 const registrationController = async (req, res) => {
   try {
     const response = await registerUser(req.body);
-    if (response.status === 200) {
-      const { status, user } = response;
-      let newUser = await User.findOne({ _id: user._id });
-      if (!newUser) {
-        newUser = new User(user);
-        await newUser.save();
-      }
-      const token = await setAuthTokens(user._id, res);
-      res.setHeader('Authorization', `Bearer ${token}`);
-      res.status(status).send({ user });
-    } else {
-      const { status, message } = response;
-      res.status(status).send({ message });
-    }
+    const { status, message } = response;
+    res.status(status).send({ message });
   } catch (err) {
     logger.error('[registrationController]', err);
     return res.status(500).json({ message: err.message });
   }
 };
 
-const getUserController = async (req, res) => {
-  return res.status(200).send(req.user);
-};
-
 const resetPasswordRequestController = async (req, res) => {
   try {
-    const resetService = await requestPasswordReset(req.body.email);
+    const resetService = await requestPasswordReset(req);
     if (resetService instanceof Error) {
       return res.status(400).json(resetService);
     } else {
@@ -77,7 +61,7 @@ const refreshController = async (req, res) => {
 
   try {
     const payload = jwt.verify(refreshToken, process.env.JWT_REFRESH_SECRET);
-    const user = await User.findOne({ _id: payload.id });
+    const user = await getUserById(payload.id, '-password -__v');
     if (!user) {
       return res.status(401).redirect('/login');
     }
@@ -86,8 +70,7 @@ const refreshController = async (req, res) => {
 
     if (process.env.NODE_ENV === 'CI') {
       const token = await setAuthTokens(userId, res);
-      const userObj = user.toJSON();
-      return res.status(200).send({ token, user: userObj });
+      return res.status(200).send({ token, user });
     }
 
     // Hash the refresh token
@@ -98,8 +81,7 @@ const refreshController = async (req, res) => {
     const session = await Session.findOne({ user: userId, refreshTokenHash: hashedToken });
     if (session && session.expiration > new Date()) {
       const token = await setAuthTokens(userId, res, session._id);
-      const userObj = user.toJSON();
-      res.status(200).send({ token, user: userObj });
+      res.status(200).send({ token, user });
     } else if (req?.query?.retry) {
       // Retrying from a refresh token request that failed (401)
       res.status(403).send('No session found');
@@ -115,7 +97,6 @@ const refreshController = async (req, res) => {
 };
 
 module.exports = {
-  getUserController,
   refreshController,
   registrationController,
   resetPasswordController,

api/server/controllers/EditController.js

@@ -112,11 +112,12 @@ const EditController = async (req, res, next, initializeClient) => {
       getReqData,
       onStart,
       abortController,
-      onProgress: progressCallback.call(null, {
+      progressCallback,
+      progressOptions: {
         res,
         text,
-        parentMessageId: overrideParentMessageId || userMessageId,
-      }),
+        // parentMessageId: overrideParentMessageId || userMessageId,
+      },
     });
 
     const conversation = await getConvo(user, conversationId);

api/server/controllers/EndpointController.js

@@ -16,10 +16,28 @@ async function endpointController(req, res) {
   /** @type {TEndpointsConfig} */
   const mergedConfig = { ...defaultEndpointsConfig, ...customConfigEndpoints };
   if (mergedConfig[EModelEndpoint.assistants] && req.app.locals?.[EModelEndpoint.assistants]) {
-    const { disableBuilder, retrievalModels, capabilities, ..._rest } =
+    const { disableBuilder, retrievalModels, capabilities, version, ..._rest } =
       req.app.locals[EModelEndpoint.assistants];
+
     mergedConfig[EModelEndpoint.assistants] = {
       ...mergedConfig[EModelEndpoint.assistants],
+      version,
+      retrievalModels,
+      disableBuilder,
+      capabilities,
+    };
+  }
+
+  if (
+    mergedConfig[EModelEndpoint.azureAssistants] &&
+    req.app.locals?.[EModelEndpoint.azureAssistants]
+  ) {
+    const { disableBuilder, retrievalModels, capabilities, version, ..._rest } =
+      req.app.locals[EModelEndpoint.azureAssistants];
+
+    mergedConfig[EModelEndpoint.azureAssistants] = {
+      ...mergedConfig[EModelEndpoint.azureAssistants],
+      version,
       retrievalModels,
       disableBuilder,
       capabilities,

api/server/controllers/PluginController.js

@@ -55,24 +55,26 @@ const getAvailablePluginsController = async (req, res) => {
       return;
     }
 
-    /** @type {{ filteredTools: string[] }} */
-    const { filteredTools = [] } = req.app.locals;
-
+    /** @type {{ filteredTools: string[], includedTools: string[] }} */
+    const { filteredTools = [], includedTools = [] } = req.app.locals;
     const pluginManifest = await fs.readFile(req.app.locals.paths.pluginManifest, 'utf8');
-
     const jsonData = JSON.parse(pluginManifest);
-    /** @type {TPlugin[]} */
+
     const uniquePlugins = filterUniquePlugins(jsonData);
-    const authenticatedPlugins = uniquePlugins.map((plugin) => {
-      if (isPluginAuthenticated(plugin)) {
-        return { ...plugin, authenticated: true };
-      } else {
-        return plugin;
-      }
-    });
+    let authenticatedPlugins = [];
+    for (const plugin of uniquePlugins) {
+      authenticatedPlugins.push(
+        isPluginAuthenticated(plugin) ? { ...plugin, authenticated: true } : plugin,
+      );
+    }
 
     let plugins = await addOpenAPISpecs(authenticatedPlugins);
-    plugins = plugins.filter((plugin) => !filteredTools.includes(plugin.pluginKey));
+
+    if (includedTools.length > 0) {
+      plugins = plugins.filter((plugin) => includedTools.includes(plugin.pluginKey));
+    } else {
+      plugins = plugins.filter((plugin) => !filteredTools.includes(plugin.pluginKey));
+    }
 
     await cache.set(CacheKeys.PLUGINS, plugins);
     res.status(200).json(plugins);

api/server/controllers/UserController.js

@@ -1,11 +1,37 @@
-const { updateUserPluginsService } = require('~/server/services/UserService');
+const {
+  Session,
+  Balance,
+  getFiles,
+  deleteFiles,
+  deleteConvos,
+  deletePresets,
+  deleteMessages,
+  deleteUserById,
+} = require('~/models');
 const { updateUserPluginAuth, deleteUserPluginAuth } = require('~/server/services/PluginService');
+const { updateUserPluginsService, deleteUserKey } = require('~/server/services/UserService');
+const { verifyEmail, resendVerificationEmail } = require('~/server/services/AuthService');
+const { processDeleteRequest } = require('~/server/services/Files/process');
+const { deleteAllSharedLinks } = require('~/models/Share');
+const { Transaction } = require('~/models/Transaction');
 const { logger } = require('~/config');
 
 const getUserController = async (req, res) => {
   res.status(200).send(req.user);
 };
 
+const deleteUserFiles = async (req) => {
+  try {
+    const userFiles = await getFiles({ user: req.user.id });
+    await processDeleteRequest({
+      req,
+      files: userFiles,
+    });
+  } catch (error) {
+    logger.error('[deleteUserFiles]', error);
+  }
+};
+
 const updateUserPluginsController = async (req, res) => {
   const { user } = req;
   const { pluginKey, action, auth, isAssistantTool } = req.body;
@@ -49,11 +75,68 @@ const updateUserPluginsController = async (req, res) => {
     res.status(200).send();
   } catch (err) {
     logger.error('[updateUserPluginsController]', err);
-    res.status(500).json({ message: err.message });
+    return res.status(500).json({ message: 'Something went wrong.' });
+  }
+};
+
+const deleteUserController = async (req, res) => {
+  const { user } = req;
+
+  try {
+    await deleteMessages({ user: user.id }); // delete user messages
+    await Session.deleteMany({ user: user.id }); // delete user sessions
+    await Transaction.deleteMany({ user: user.id }); // delete user transactions
+    await deleteUserKey({ userId: user.id, all: true }); // delete user keys
+    await Balance.deleteMany({ user: user._id }); // delete user balances
+    await deletePresets(user.id); // delete user presets
+    /* TODO: Delete Assistant Threads */
+    await deleteConvos(user.id); // delete user convos
+    await deleteUserPluginAuth(user.id, null, true); // delete user plugin auth
+    await deleteUserById(user.id); // delete user
+    await deleteAllSharedLinks(user.id); // delete user shared links
+    await deleteUserFiles(req); // delete user files
+    await deleteFiles(null, user.id); // delete database files in case of orphaned files from previous steps
+    /* TODO: queue job for cleaning actions and assistants of non-existant users */
+    logger.info(`User deleted account. Email: ${user.email} ID: ${user.id}`);
+    res.status(200).send({ message: 'User deleted' });
+  } catch (err) {
+    logger.error('[deleteUserController]', err);
+    return res.status(500).json({ message: 'Something went wrong.' });
+  }
+};
+
+const verifyEmailController = async (req, res) => {
+  try {
+    const verifyEmailService = await verifyEmail(req);
+    if (verifyEmailService instanceof Error) {
+      return res.status(400).json(verifyEmailService);
+    } else {
+      return res.status(200).json(verifyEmailService);
+    }
+  } catch (e) {
+    logger.error('[verifyEmailController]', e);
+    return res.status(500).json({ message: 'Something went wrong.' });
+  }
+};
+
+const resendVerificationController = async (req, res) => {
+  try {
+    const result = await resendVerificationEmail(req);
+    if (result instanceof Error) {
+      return res.status(400).json(result);
+    } else {
+      return res.status(200).json(result);
+    }
+  } catch (e) {
+    logger.error('[verifyEmailController]', e);
+    return res.status(500).json({ message: 'Something went wrong.' });
   }
 };
 
 module.exports = {
   getUserController,
+  deleteUserController,
+  verifyEmailController,
   updateUserPluginsController,
+  resendVerificationController,
 };

api/server/controllers/assistants/chatV1.js

@@ -1,14 +1,13 @@
 const { v4 } = require('uuid');
-const express = require('express');
 const {
   Constants,
   RunStatus,
   CacheKeys,
-  FileSources,
   ContentTypes,
   EModelEndpoint,
   ViolationTypes,
   ImageVisionTool,
+  checkOpenAIStorage,
   AssistantStreamEvents,
 } = require('librechat-data-provider');
 const {
@@ -21,44 +20,36 @@ const {
 } = require('~/server/services/Threads');
 const { sendResponse, sendMessage, sleep, isEnabled, countTokens } = require('~/server/utils');
 const { runAssistant, createOnTextProgress } = require('~/server/services/AssistantService');
-const { addTitle, initializeClient } = require('~/server/services/Endpoints/assistants');
+const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
 const { formatMessage, createVisionPrompt } = require('~/app/clients/prompts');
 const { createRun, StreamRunManager } = require('~/server/services/Runs');
+const { addTitle } = require('~/server/services/Endpoints/assistants');
 const { getTransactions } = require('~/models/Transaction');
 const checkBalance = require('~/models/checkBalance');
 const { getConvo } = require('~/models/Conversation');
 const getLogStores = require('~/cache/getLogStores');
 const { getModelMaxTokens } = require('~/utils');
+const { getOpenAIClient } = require('./helpers');
 const { logger } = require('~/config');
 
-const router = express.Router();
-const {
-  setHeaders,
-  handleAbort,
-  validateModel,
-  handleAbortError,
-  // validateEndpoint,
-  buildEndpointOption,
-} = require('~/server/middleware');
-
-router.post('/abort', handleAbort());
-
 const ten_minutes = 1000 * 60 * 10;
 
 /**
  * @route POST /
  * @desc Chat with an assistant
  * @access Public
- * @param {express.Request} req - The request object, containing the request data.
- * @param {express.Response} res - The response object, used to send back a response.
+ * @param {object} req - The request object, containing the request data.
+ * @param {object} req.body - The request payload.
+ * @param {Express.Response} res - The response object, used to send back a response.
  * @returns {void}
  */
-router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res) => {
+const chatV1 = async (req, res) => {
   logger.debug('[/assistants/chat/] req.body', req.body);
 
   const {
     text,
     model,
+    endpoint,
     files = [],
     promptPrefix,
     assistant_id,
@@ -69,30 +60,6 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
     parentMessageId: _parentId = Constants.NO_PARENT,
   } = req.body;
 
-  /** @type {Partial<TAssistantEndpoint>} */
-  const assistantsConfig = req.app.locals?.[EModelEndpoint.assistants];
-
-  if (assistantsConfig) {
-    const { supportedIds, excludedIds } = assistantsConfig;
-    const error = { message: 'Assistant not supported' };
-    if (supportedIds?.length && !supportedIds.includes(assistant_id)) {
-      return await handleAbortError(res, req, error, {
-        sender: 'System',
-        conversationId: convoId,
-        messageId: v4(),
-        parentMessageId: _messageId,
-        error,
-      });
-    } else if (excludedIds?.length && excludedIds.includes(assistant_id)) {
-      return await handleAbortError(res, req, error, {
-        sender: 'System',
-        conversationId: convoId,
-        messageId: v4(),
-        parentMessageId: _messageId,
-      });
-    }
-  }
-
   /** @type {OpenAIClient} */
   let openai;
   /** @type {string|undefined} - the current thread id */
@@ -138,7 +105,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
       user: req.user.id,
       shouldSaveMessage: false,
       messageId: responseMessageId,
-      endpoint: EModelEndpoint.assistants,
+      endpoint,
     };
 
     if (error.message === 'Run cancelled') {
@@ -149,7 +116,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
       logger.debug('[/assistants/chat/] Request aborted on close');
     } else if (/Files.*are invalid/.test(error.message)) {
       const errorMessage = `Files are invalid, or may not have uploaded yet.${
-        req.app.locals?.[EModelEndpoint.azureOpenAI].assistants
+        endpoint === EModelEndpoint.azureAssistants
           ? ' If using Azure OpenAI, files are only available in the region of the assistant\'s model at the time of upload.'
           : ''
       }`;
@@ -205,6 +172,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
       const runMessages = await checkMessageGaps({
         openai,
         run_id,
+        endpoint,
         thread_id,
         conversationId,
         latestMessageId: responseMessageId,
@@ -311,8 +279,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
       });
     };
 
-    /** @type {{ openai: OpenAIClient }} */
-    const { openai: _openai, client } = await initializeClient({
+    const { openai: _openai, client } = await getOpenAIClient({
       req,
       res,
       endpointOption: req.body.endpointOption,
@@ -320,6 +287,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
     });
 
     openai = _openai;
+    await validateAuthor({ req, openai });
 
     if (previousMessages.length) {
       parentMessageId = previousMessages[previousMessages.length - 1].messageId;
@@ -370,10 +338,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
 
       /** @type {MongoFile[]} */
       const attachments = await req.body.endpointOption.attachments;
-      if (
-        attachments &&
-        attachments.every((attachment) => attachment.source === FileSources.openai)
-      ) {
+      if (attachments && attachments.every((attachment) => checkOpenAIStorage(attachment.source))) {
         return;
       }
 
@@ -431,7 +396,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
 
       if (processedFiles) {
         for (const file of processedFiles) {
-          if (file.source !== FileSources.openai) {
+          if (!checkOpenAIStorage(file.source)) {
             attachedFileIds.delete(file.file_id);
             const index = file_ids.indexOf(file.file_id);
             if (index > -1) {
@@ -467,6 +432,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
         assistant_id,
         thread_id,
         model: assistant_id,
+        endpoint,
       };
 
       previousMessages.push(requestMessage);
@@ -476,7 +442,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
 
       conversation = {
         conversationId,
-        endpoint: EModelEndpoint.assistants,
+        endpoint,
         promptPrefix: promptPrefix,
         instructions: instructions,
         assistant_id,
@@ -513,7 +479,8 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
     let response;
 
     const processRun = async (retry = false) => {
-      if (req.app.locals[EModelEndpoint.azureOpenAI]?.assistants) {
+      if (endpoint === EModelEndpoint.azureAssistants) {
+        body.model = openai._options.model;
         openai.attachedFileIds = attachedFileIds;
         openai.visionPromise = visionPromise;
         if (retry) {
@@ -602,6 +569,7 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
       assistant_id,
       thread_id,
       model: assistant_id,
+      endpoint,
     };
 
     sendMessage(res, {
@@ -654,6 +622,6 @@ router.post('/', validateModel, buildEndpointOption, setHeaders, async (req, res
   } catch (error) {
     await handleError(error);
   }
-});
+};
 
-module.exports = router;
+module.exports = chatV1;

api/server/controllers/assistants/chatV2.js

@@ -0,0 +1,597 @@
+const { v4 } = require('uuid');
+const {
+  Constants,
+  RunStatus,
+  CacheKeys,
+  ContentTypes,
+  ToolCallTypes,
+  EModelEndpoint,
+  ViolationTypes,
+  retrievalMimeTypes,
+  AssistantStreamEvents,
+} = require('librechat-data-provider');
+const {
+  initThread,
+  recordUsage,
+  saveUserMessage,
+  checkMessageGaps,
+  addThreadMetadata,
+  saveAssistantMessage,
+} = require('~/server/services/Threads');
+const { sendResponse, sendMessage, sleep, isEnabled, countTokens } = require('~/server/utils');
+const { runAssistant, createOnTextProgress } = require('~/server/services/AssistantService');
+const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
+const { createRun, StreamRunManager } = require('~/server/services/Runs');
+const { addTitle } = require('~/server/services/Endpoints/assistants');
+const { getTransactions } = require('~/models/Transaction');
+const checkBalance = require('~/models/checkBalance');
+const { getConvo } = require('~/models/Conversation');
+const getLogStores = require('~/cache/getLogStores');
+const { getModelMaxTokens } = require('~/utils');
+const { getOpenAIClient } = require('./helpers');
+const { logger } = require('~/config');
+
+const ten_minutes = 1000 * 60 * 10;
+
+/**
+ * @route POST /
+ * @desc Chat with an assistant
+ * @access Public
+ * @param {Express.Request} req - The request object, containing the request data.
+ * @param {Express.Response} res - The response object, used to send back a response.
+ * @returns {void}
+ */
+const chatV2 = async (req, res) => {
+  logger.debug('[/assistants/chat/] req.body', req.body);
+
+  /** @type {{ files: MongoFile[]}} */
+  const {
+    text,
+    model,
+    endpoint,
+    files = [],
+    promptPrefix,
+    assistant_id,
+    instructions,
+    thread_id: _thread_id,
+    messageId: _messageId,
+    conversationId: convoId,
+    parentMessageId: _parentId = Constants.NO_PARENT,
+  } = req.body;
+
+  /** @type {OpenAIClient} */
+  let openai;
+  /** @type {string|undefined} - the current thread id */
+  let thread_id = _thread_id;
+  /** @type {string|undefined} - the current run id */
+  let run_id;
+  /** @type {string|undefined} - the parent messageId */
+  let parentMessageId = _parentId;
+  /** @type {TMessage[]} */
+  let previousMessages = [];
+  /** @type {import('librechat-data-provider').TConversation | null} */
+  let conversation = null;
+  /** @type {string[]} */
+  let file_ids = [];
+  /** @type {Set<string>} */
+  let attachedFileIds = new Set();
+  /** @type {TMessage | null} */
+  let requestMessage = null;
+
+  const userMessageId = v4();
+  const responseMessageId = v4();
+
+  /** @type {string} - The conversation UUID - created if undefined */
+  const conversationId = convoId ?? v4();
+
+  const cache = getLogStores(CacheKeys.ABORT_KEYS);
+  const cacheKey = `${req.user.id}:${conversationId}`;
+
+  /** @type {Run | undefined} - The completed run, undefined if incomplete */
+  let completedRun;
+
+  const handleError = async (error) => {
+    const defaultErrorMessage =
+      'The Assistant run failed to initialize. Try sending a message in a new conversation.';
+    const messageData = {
+      thread_id,
+      assistant_id,
+      conversationId,
+      parentMessageId,
+      sender: 'System',
+      user: req.user.id,
+      shouldSaveMessage: false,
+      messageId: responseMessageId,
+      endpoint,
+    };
+
+    if (error.message === 'Run cancelled') {
+      return res.end();
+    } else if (error.message === 'Request closed' && completedRun) {
+      return;
+    } else if (error.message === 'Request closed') {
+      logger.debug('[/assistants/chat/] Request aborted on close');
+    } else if (/Files.*are invalid/.test(error.message)) {
+      const errorMessage = `Files are invalid, or may not have uploaded yet.${
+        endpoint === EModelEndpoint.azureAssistants
+          ? ' If using Azure OpenAI, files are only available in the region of the assistant\'s model at the time of upload.'
+          : ''
+      }`;
+      return sendResponse(res, messageData, errorMessage);
+    } else if (error?.message?.includes('string too long')) {
+      return sendResponse(
+        res,
+        messageData,
+        'Message too long. The Assistants API has a limit of 32,768 characters per message. Please shorten it and try again.',
+      );
+    } else if (error?.message?.includes(ViolationTypes.TOKEN_BALANCE)) {
+      return sendResponse(res, messageData, error.message);
+    } else {
+      logger.error('[/assistants/chat/]', error);
+    }
+
+    if (!openai || !thread_id || !run_id) {
+      return sendResponse(res, messageData, defaultErrorMessage);
+    }
+
+    await sleep(2000);
+
+    try {
+      const status = await cache.get(cacheKey);
+      if (status === 'cancelled') {
+        logger.debug('[/assistants/chat/] Run already cancelled');
+        return res.end();
+      }
+      await cache.delete(cacheKey);
+      const cancelledRun = await openai.beta.threads.runs.cancel(thread_id, run_id);
+      logger.debug('[/assistants/chat/] Cancelled run:', cancelledRun);
+    } catch (error) {
+      logger.error('[/assistants/chat/] Error cancelling run', error);
+    }
+
+    await sleep(2000);
+
+    let run;
+    try {
+      run = await openai.beta.threads.runs.retrieve(thread_id, run_id);
+      await recordUsage({
+        ...run.usage,
+        model: run.model,
+        user: req.user.id,
+        conversationId,
+      });
+    } catch (error) {
+      logger.error('[/assistants/chat/] Error fetching or processing run', error);
+    }
+
+    let finalEvent;
+    try {
+      const runMessages = await checkMessageGaps({
+        openai,
+        run_id,
+        endpoint,
+        thread_id,
+        conversationId,
+        latestMessageId: responseMessageId,
+      });
+
+      const errorContentPart = {
+        text: {
+          value:
+            error?.message ?? 'There was an error processing your request. Please try again later.',
+        },
+        type: ContentTypes.ERROR,
+      };
+
+      if (!Array.isArray(runMessages[runMessages.length - 1]?.content)) {
+        runMessages[runMessages.length - 1].content = [errorContentPart];
+      } else {
+        const contentParts = runMessages[runMessages.length - 1].content;
+        for (let i = 0; i < contentParts.length; i++) {
+          const currentPart = contentParts[i];
+          /** @type {CodeToolCall | RetrievalToolCall | FunctionToolCall | undefined} */
+          const toolCall = currentPart?.[ContentTypes.TOOL_CALL];
+          if (
+            toolCall &&
+            toolCall?.function &&
+            !(toolCall?.function?.output || toolCall?.function?.output?.length)
+          ) {
+            contentParts[i] = {
+              ...currentPart,
+              [ContentTypes.TOOL_CALL]: {
+                ...toolCall,
+                function: {
+                  ...toolCall.function,
+                  output: 'error processing tool',
+                },
+              },
+            };
+          }
+        }
+        runMessages[runMessages.length - 1].content.push(errorContentPart);
+      }
+
+      finalEvent = {
+        final: true,
+        conversation: await getConvo(req.user.id, conversationId),
+        runMessages,
+      };
+    } catch (error) {
+      logger.error('[/assistants/chat/] Error finalizing error process', error);
+      return sendResponse(res, messageData, 'The Assistant run failed');
+    }
+
+    return sendResponse(res, finalEvent);
+  };
+
+  try {
+    res.on('close', async () => {
+      if (!completedRun) {
+        await handleError(new Error('Request closed'));
+      }
+    });
+
+    if (convoId && !_thread_id) {
+      completedRun = true;
+      throw new Error('Missing thread_id for existing conversation');
+    }
+
+    if (!assistant_id) {
+      completedRun = true;
+      throw new Error('Missing assistant_id');
+    }
+
+    const checkBalanceBeforeRun = async () => {
+      if (!isEnabled(process.env.CHECK_BALANCE)) {
+        return;
+      }
+      const transactions =
+        (await getTransactions({
+          user: req.user.id,
+          context: 'message',
+          conversationId,
+        })) ?? [];
+
+      const totalPreviousTokens = Math.abs(
+        transactions.reduce((acc, curr) => acc + curr.rawAmount, 0),
+      );
+
+      // TODO: make promptBuffer a config option; buffer for titles, needs buffer for system instructions
+      const promptBuffer = parentMessageId === Constants.NO_PARENT && !_thread_id ? 200 : 0;
+      // 5 is added for labels
+      let promptTokens = (await countTokens(text + (promptPrefix ?? ''))) + 5;
+      promptTokens += totalPreviousTokens + promptBuffer;
+      // Count tokens up to the current context window
+      promptTokens = Math.min(promptTokens, getModelMaxTokens(model));
+
+      await checkBalance({
+        req,
+        res,
+        txData: {
+          model,
+          user: req.user.id,
+          tokenType: 'prompt',
+          amount: promptTokens,
+        },
+      });
+    };
+
+    const { openai: _openai, client } = await getOpenAIClient({
+      req,
+      res,
+      endpointOption: req.body.endpointOption,
+      initAppClient: true,
+    });
+
+    openai = _openai;
+    await validateAuthor({ req, openai });
+
+    if (previousMessages.length) {
+      parentMessageId = previousMessages[previousMessages.length - 1].messageId;
+    }
+
+    let userMessage = {
+      role: 'user',
+      content: [
+        {
+          type: ContentTypes.TEXT,
+          text,
+        },
+      ],
+      metadata: {
+        messageId: userMessageId,
+      },
+    };
+
+    /** @type {CreateRunBody | undefined} */
+    const body = {
+      assistant_id,
+      model,
+    };
+
+    if (promptPrefix) {
+      body.additional_instructions = promptPrefix;
+    }
+
+    if (instructions) {
+      body.instructions = instructions;
+    }
+
+    const getRequestFileIds = async () => {
+      let thread_file_ids = [];
+      if (convoId) {
+        const convo = await getConvo(req.user.id, convoId);
+        if (convo && convo.file_ids) {
+          thread_file_ids = convo.file_ids;
+        }
+      }
+
+      if (files.length || thread_file_ids.length) {
+        attachedFileIds = new Set([...file_ids, ...thread_file_ids]);
+
+        let attachmentIndex = 0;
+        for (const file of files) {
+          file_ids.push(file.file_id);
+          if (file.type.startsWith('image')) {
+            userMessage.content.push({
+              type: ContentTypes.IMAGE_FILE,
+              [ContentTypes.IMAGE_FILE]: { file_id: file.file_id },
+            });
+          }
+
+          if (!userMessage.attachments) {
+            userMessage.attachments = [];
+          }
+
+          userMessage.attachments.push({
+            file_id: file.file_id,
+            tools: [{ type: ToolCallTypes.CODE_INTERPRETER }],
+          });
+
+          if (file.type.startsWith('image')) {
+            continue;
+          }
+
+          const mimeType = file.type;
+          const isSupportedByRetrieval = retrievalMimeTypes.some((regex) => regex.test(mimeType));
+          if (isSupportedByRetrieval) {
+            userMessage.attachments[attachmentIndex].tools.push({
+              type: ToolCallTypes.FILE_SEARCH,
+            });
+          }
+
+          attachmentIndex++;
+        }
+      }
+    };
+
+    const initializeThread = async () => {
+      await getRequestFileIds();
+
+      // TODO: may allow multiple messages to be created beforehand in a future update
+      const initThreadBody = {
+        messages: [userMessage],
+        metadata: {
+          user: req.user.id,
+          conversationId,
+        },
+      };
+
+      const result = await initThread({ openai, body: initThreadBody, thread_id });
+      thread_id = result.thread_id;
+
+      createOnTextProgress({
+        openai,
+        conversationId,
+        userMessageId,
+        messageId: responseMessageId,
+        thread_id,
+      });
+
+      requestMessage = {
+        user: req.user.id,
+        text,
+        messageId: userMessageId,
+        parentMessageId,
+        // TODO: make sure client sends correct format for `files`, use zod
+        files,
+        file_ids,
+        conversationId,
+        isCreatedByUser: true,
+        assistant_id,
+        thread_id,
+        model: assistant_id,
+        endpoint,
+      };
+
+      previousMessages.push(requestMessage);
+
+      /* asynchronous */
+      saveUserMessage({ ...requestMessage, model });
+
+      conversation = {
+        conversationId,
+        endpoint,
+        promptPrefix: promptPrefix,
+        instructions: instructions,
+        assistant_id,
+        // model,
+      };
+
+      if (file_ids.length) {
+        conversation.file_ids = file_ids;
+      }
+    };
+
+    const promises = [initializeThread(), checkBalanceBeforeRun()];
+    await Promise.all(promises);
+
+    const sendInitialResponse = () => {
+      sendMessage(res, {
+        sync: true,
+        conversationId,
+        // messages: previousMessages,
+        requestMessage,
+        responseMessage: {
+          user: req.user.id,
+          messageId: openai.responseMessage.messageId,
+          parentMessageId: userMessageId,
+          conversationId,
+          assistant_id,
+          thread_id,
+          model: assistant_id,
+        },
+      });
+    };
+
+    /** @type {RunResponse | typeof StreamRunManager | undefined} */
+    let response;
+
+    const processRun = async (retry = false) => {
+      if (endpoint === EModelEndpoint.azureAssistants) {
+        body.model = openai._options.model;
+        openai.attachedFileIds = attachedFileIds;
+        if (retry) {
+          response = await runAssistant({
+            openai,
+            thread_id,
+            run_id,
+            in_progress: openai.in_progress,
+          });
+          return;
+        }
+
+        /* NOTE:
+         * By default, a Run will use the model and tools configuration specified in Assistant object,
+         * but you can override most of these when creating the Run for added flexibility:
+         */
+        const run = await createRun({
+          openai,
+          thread_id,
+          body,
+        });
+
+        run_id = run.id;
+        await cache.set(cacheKey, `${thread_id}:${run_id}`, ten_minutes);
+        sendInitialResponse();
+
+        // todo: retry logic
+        response = await runAssistant({ openai, thread_id, run_id });
+        return;
+      }
+
+      /** @type {{[AssistantStreamEvents.ThreadRunCreated]: (event: ThreadRunCreated) => Promise<void>}} */
+      const handlers = {
+        [AssistantStreamEvents.ThreadRunCreated]: async (event) => {
+          await cache.set(cacheKey, `${thread_id}:${event.data.id}`, ten_minutes);
+          run_id = event.data.id;
+          sendInitialResponse();
+        },
+      };
+
+      const streamRunManager = new StreamRunManager({
+        req,
+        res,
+        openai,
+        handlers,
+        thread_id,
+        attachedFileIds,
+        parentMessageId: userMessageId,
+        responseMessage: openai.responseMessage,
+        // streamOptions: {
+
+        // },
+      });
+
+      await streamRunManager.runAssistant({
+        thread_id,
+        body,
+      });
+
+      response = streamRunManager;
+      response.text = streamRunManager.intermediateText;
+    };
+
+    await processRun();
+    logger.debug('[/assistants/chat/] response', {
+      run: response.run,
+      steps: response.steps,
+    });
+
+    if (response.run.status === RunStatus.CANCELLED) {
+      logger.debug('[/assistants/chat/] Run cancelled, handled by `abortRun`');
+      return res.end();
+    }
+
+    if (response.run.status === RunStatus.IN_PROGRESS) {
+      processRun(true);
+    }
+
+    completedRun = response.run;
+
+    /** @type {ResponseMessage} */
+    const responseMessage = {
+      ...(response.responseMessage ?? response.finalMessage),
+      text: response.text,
+      parentMessageId: userMessageId,
+      conversationId,
+      user: req.user.id,
+      assistant_id,
+      thread_id,
+      model: assistant_id,
+      endpoint,
+    };
+
+    sendMessage(res, {
+      final: true,
+      conversation,
+      requestMessage: {
+        parentMessageId,
+        thread_id,
+      },
+    });
+    res.end();
+
+    await saveAssistantMessage({ ...responseMessage, model });
+
+    if (parentMessageId === Constants.NO_PARENT && !_thread_id) {
+      addTitle(req, {
+        text,
+        responseText: response.text,
+        conversationId,
+        client,
+      });
+    }
+
+    await addThreadMetadata({
+      openai,
+      thread_id,
+      messageId: responseMessage.messageId,
+      messages: response.messages,
+    });
+
+    if (!response.run.usage) {
+      await sleep(3000);
+      completedRun = await openai.beta.threads.runs.retrieve(thread_id, response.run.id);
+      if (completedRun.usage) {
+        await recordUsage({
+          ...completedRun.usage,
+          user: req.user.id,
+          model: completedRun.model ?? model,
+          conversationId,
+        });
+      }
+    } else {
+      await recordUsage({
+        ...response.run.usage,
+        user: req.user.id,
+        model: response.run.model ?? model,
+        conversationId,
+      });
+    }
+  } catch (error) {
+    await handleError(error);
+  }
+};
+
+module.exports = chatV2;

api/server/controllers/assistants/helpers.js

@@ -0,0 +1,269 @@
+const {
+  EModelEndpoint,
+  CacheKeys,
+  defaultAssistantsVersion,
+  defaultOrderQuery,
+} = require('librechat-data-provider');
+const {
+  initializeClient: initAzureClient,
+} = require('~/server/services/Endpoints/azureAssistants');
+const { initializeClient } = require('~/server/services/Endpoints/assistants');
+const { getLogStores } = require('~/cache');
+
+/**
+ * @param {Express.Request} req
+ * @param {string} [endpoint]
+ * @returns {Promise<string>}
+ */
+const getCurrentVersion = async (req, endpoint) => {
+  const index = req.baseUrl.lastIndexOf('/v');
+  let version = index !== -1 ? req.baseUrl.substring(index + 1, index + 3) : null;
+  if (!version && req.body.version) {
+    version = `v${req.body.version}`;
+  }
+  if (!version && endpoint) {
+    const cache = getLogStores(CacheKeys.CONFIG_STORE);
+    const cachedEndpointsConfig = await cache.get(CacheKeys.ENDPOINT_CONFIG);
+    version = `v${
+      cachedEndpointsConfig?.[endpoint]?.version ?? defaultAssistantsVersion[endpoint]
+    }`;
+  }
+  if (!version?.startsWith('v') && version.length !== 2) {
+    throw new Error(`[${req.baseUrl}] Invalid version: ${version}`);
+  }
+  return version;
+};
+
+/**
+ * Asynchronously lists assistants based on provided query parameters.
+ *
+ * Initializes the client with the current request and response objects and lists assistants
+ * according to the query parameters. This function abstracts the logic for non-Azure paths.
+ *
+ * @deprecated
+ * @async
+ * @param {object} params - The parameters object.
+ * @param {object} params.req - The request object, used for initializing the client.
+ * @param {object} params.res - The response object, used for initializing the client.
+ * @param {string} params.version - The API version to use.
+ * @param {object} params.query - The query parameters to list assistants (e.g., limit, order).
+ * @returns {Promise<object>} A promise that resolves to the response from the `openai.beta.assistants.list` method call.
+ */
+const _listAssistants = async ({ req, res, version, query }) => {
+  const { openai } = await getOpenAIClient({ req, res, version });
+  return openai.beta.assistants.list(query);
+};
+
+/**
+ * Fetches all assistants based on provided query params, until `has_more` is `false`.
+ *
+ * @async
+ * @param {object} params - The parameters object.
+ * @param {object} params.req - The request object, used for initializing the client.
+ * @param {object} params.res - The response object, used for initializing the client.
+ * @param {string} params.version - The API version to use.
+ * @param {Omit<AssistantListParams, 'endpoint'>} params.query - The query parameters to list assistants (e.g., limit, order).
+ * @returns {Promise<object>} A promise that resolves to the response from the `openai.beta.assistants.list` method call.
+ */
+const listAllAssistants = async ({ req, res, version, query }) => {
+  /** @type {{ openai: OpenAIClient }} */
+  const { openai } = await getOpenAIClient({ req, res, version });
+  const allAssistants = [];
+
+  let first_id;
+  let last_id;
+  let afterToken = query.after;
+  let hasMore = true;
+
+  while (hasMore) {
+    const response = await openai.beta.assistants.list({
+      ...query,
+      after: afterToken,
+    });
+
+    const { body } = response;
+
+    allAssistants.push(...body.data);
+    hasMore = body.has_more;
+
+    if (!first_id) {
+      first_id = body.first_id;
+    }
+
+    if (hasMore) {
+      afterToken = body.last_id;
+    } else {
+      last_id = body.last_id;
+    }
+  }
+
+  return {
+    data: allAssistants,
+    body: {
+      data: allAssistants,
+      has_more: false,
+      first_id,
+      last_id,
+    },
+  };
+};
+
+/**
+ * Asynchronously lists assistants for Azure configured groups.
+ *
+ * Iterates through Azure configured assistant groups, initializes the client with the current request and response objects,
+ * lists assistants based on the provided query parameters, and merges their data alongside the model information into a single array.
+ *
+ * @async
+ * @param {object} params - The parameters object.
+ * @param {object} params.req - The request object, used for initializing the client and manipulating the request body.
+ * @param {object} params.res - The response object, used for initializing the client.
+ * @param {string} params.version - The API version to use.
+ * @param {TAzureConfig} params.azureConfig - The Azure configuration object containing assistantGroups and groupMap.
+ * @param {object} params.query - The query parameters to list assistants (e.g., limit, order).
+ * @returns {Promise<AssistantListResponse>} A promise that resolves to an array of assistant data merged with their respective model information.
+ */
+const listAssistantsForAzure = async ({ req, res, version, azureConfig = {}, query }) => {
+  /** @type {Array<[string, TAzureModelConfig]>} */
+  const groupModelTuples = [];
+  const promises = [];
+  /** @type {Array<TAzureGroup>} */
+  const groups = [];
+
+  const { groupMap, assistantGroups } = azureConfig;
+
+  for (const groupName of assistantGroups) {
+    const group = groupMap[groupName];
+    groups.push(group);
+
+    const currentModelTuples = Object.entries(group?.models);
+    groupModelTuples.push(currentModelTuples);
+
+    /* The specified model is only necessary to
+    fetch assistants for the shared instance */
+    req.body.model = currentModelTuples[0][0];
+    promises.push(listAllAssistants({ req, res, version, query }));
+  }
+
+  const resolvedQueries = await Promise.all(promises);
+  const data = resolvedQueries.flatMap((res, i) =>
+    res.data.map((assistant) => {
+      const deploymentName = assistant.model;
+      const currentGroup = groups[i];
+      const currentModelTuples = groupModelTuples[i];
+      const firstModel = currentModelTuples[0][0];
+
+      if (currentGroup.deploymentName === deploymentName) {
+        return { ...assistant, model: firstModel };
+      }
+
+      for (const [model, modelConfig] of currentModelTuples) {
+        if (modelConfig.deploymentName === deploymentName) {
+          return { ...assistant, model };
+        }
+      }
+
+      return { ...assistant, model: firstModel };
+    }),
+  );
+
+  return {
+    first_id: data[0]?.id,
+    last_id: data[data.length - 1]?.id,
+    object: 'list',
+    has_more: false,
+    data,
+  };
+};
+
+async function getOpenAIClient({ req, res, endpointOption, initAppClient, overrideEndpoint }) {
+  let endpoint = overrideEndpoint ?? req.body.endpoint ?? req.query.endpoint;
+  const version = await getCurrentVersion(req, endpoint);
+  if (!endpoint) {
+    throw new Error(`[${req.baseUrl}] Endpoint is required`);
+  }
+
+  let result;
+  if (endpoint === EModelEndpoint.assistants) {
+    result = await initializeClient({ req, res, version, endpointOption, initAppClient });
+  } else if (endpoint === EModelEndpoint.azureAssistants) {
+    result = await initAzureClient({ req, res, version, endpointOption, initAppClient });
+  }
+
+  return result;
+}
+
+/**
+ * Returns a list of assistants.
+ * @param {object} params
+ * @param {object} params.req - Express Request
+ * @param {AssistantListParams} [params.req.query] - The assistant list parameters for pagination and sorting.
+ * @param {object} params.res - Express Response
+ * @param {string} [params.overrideEndpoint] - The endpoint to override the request endpoint.
+ * @returns {Promise<AssistantListResponse>} 200 - success response - application/json
+ */
+const fetchAssistants = async ({ req, res, overrideEndpoint }) => {
+  const {
+    limit = 100,
+    order = 'desc',
+    after,
+    before,
+    endpoint,
+  } = req.query ?? {
+    endpoint: overrideEndpoint,
+    ...defaultOrderQuery,
+  };
+
+  const version = await getCurrentVersion(req, endpoint);
+  const query = { limit, order, after, before };
+
+  /** @type {AssistantListResponse} */
+  let body;
+
+  if (endpoint === EModelEndpoint.assistants) {
+    ({ body } = await listAllAssistants({ req, res, version, query }));
+  } else if (endpoint === EModelEndpoint.azureAssistants) {
+    const azureConfig = req.app.locals[EModelEndpoint.azureOpenAI];
+    body = await listAssistantsForAzure({ req, res, version, azureConfig, query });
+  }
+
+  if (req.user.role === 'ADMIN') {
+    return body;
+  } else if (!req.app.locals[endpoint]) {
+    return body;
+  }
+
+  body.data = filterAssistants({
+    userId: req.user.id,
+    assistants: body.data,
+    assistantsConfig: req.app.locals[endpoint],
+  });
+  return body;
+};
+
+/**
+ * Filter assistants based on configuration.
+ *
+ * @param {object} params - The parameters object.
+ * @param {string} params.userId -  The user ID to filter private assistants.
+ * @param {Assistant[]} params.assistants - The list of assistants to filter.
+ * @param {Partial<TAssistantEndpoint>} params.assistantsConfig -  The assistant configuration.
+ * @returns {Assistant[]} - The filtered list of assistants.
+ */
+function filterAssistants({ assistants, userId, assistantsConfig }) {
+  const { supportedIds, excludedIds, privateAssistants } = assistantsConfig;
+  if (privateAssistants) {
+    return assistants.filter((assistant) => userId === assistant.metadata?.author);
+  } else if (supportedIds?.length) {
+    return assistants.filter((assistant) => supportedIds.includes(assistant.id));
+  } else if (excludedIds?.length) {
+    return assistants.filter((assistant) => !excludedIds.includes(assistant.id));
+  }
+  return assistants;
+}
+
+module.exports = {
+  getOpenAIClient,
+  fetchAssistants,
+  getCurrentVersion,
+};

api/server/controllers/assistants/v1.js

@@ -1,34 +1,12 @@
-const multer = require('multer');
-const express = require('express');
-const { FileContext, EModelEndpoint } = require('librechat-data-provider');
-const {
-  initializeClient,
-  listAssistantsForAzure,
-  listAssistants,
-} = require('~/server/services/Endpoints/assistants');
+const { FileContext } = require('librechat-data-provider');
+const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
+const { deleteAssistantActions } = require('~/server/services/ActionService');
+const { updateAssistantDoc, getAssistants } = require('~/models/Assistant');
 const { uploadImageBuffer } = require('~/server/services/Files/process');
-const { updateAssistant, getAssistants } = require('~/models/Assistant');
+const { getOpenAIClient, fetchAssistants } = require('./helpers');
 const { deleteFileByFilter } = require('~/models/File');
 const { logger } = require('~/config');
-const actions = require('./actions');
-const tools = require('./tools');
-
-const upload = multer();
-const router = express.Router();
-
-/**
- * Assistant actions route.
- * @route GET|POST /assistants/actions
- */
-router.use('/actions', actions);
-
-/**
- * Create an assistant.
- * @route GET /assistants/tools
- * @returns {TPlugin[]} 200 - application/json
- */
-router.use('/tools', tools);
 
 /**
  * Create an assistant.
@@ -36,12 +14,11 @@ router.use('/tools', tools);
  * @param {AssistantCreateParams} req.body - The assistant creation parameters.
  * @returns {Assistant} 201 - success response - application/json
  */
-router.post('/', async (req, res) => {
+const createAssistant = async (req, res) => {
   try {
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
 
-    const { tools = [], ...assistantData } = req.body;
+    const { tools = [], endpoint, ...assistantData } = req.body;
     assistantData.tools = tools
       .map((tool) => {
         if (typeof tool !== 'string') {
@@ -52,18 +29,30 @@ router.post('/', async (req, res) => {
       })
       .filter((tool) => tool);
 
+    let azureModelIdentifier = null;
     if (openai.locals?.azureOptions) {
+      azureModelIdentifier = assistantData.model;
       assistantData.model = openai.locals.azureOptions.azureOpenAIApiDeploymentName;
     }
 
+    assistantData.metadata = {
+      author: req.user.id,
+      endpoint,
+    };
+
     const assistant = await openai.beta.assistants.create(assistantData);
+    const promise = updateAssistantDoc({ assistant_id: assistant.id }, { user: req.user.id });
+    if (azureModelIdentifier) {
+      assistant.model = azureModelIdentifier;
+    }
+    await promise;
     logger.debug('/assistants/', assistant);
     res.status(201).json(assistant);
   } catch (error) {
     logger.error('[/assistants] Error creating assistant', error);
     res.status(500).json({ error: error.message });
   }
-});
+};
 
 /**
  * Retrieves an assistant.
@@ -71,11 +60,10 @@ router.post('/', async (req, res) => {
  * @param {string} req.params.id - Assistant identifier.
  * @returns {Assistant} 200 - success response - application/json
  */
-router.get('/:id', async (req, res) => {
+const retrieveAssistant = async (req, res) => {
   try {
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
-
+    /* NOTE: not actually being used right now */
+    const { openai } = await getOpenAIClient({ req, res });
     const assistant_id = req.params.id;
     const assistant = await openai.beta.assistants.retrieve(assistant_id);
     res.json(assistant);
@@ -83,22 +71,24 @@ router.get('/:id', async (req, res) => {
     logger.error('[/assistants/:id] Error retrieving assistant', error);
     res.status(500).json({ error: error.message });
   }
-});
+};
 
 /**
  * Modifies an assistant.
  * @route PATCH /assistants/:id
+ * @param {object} req - Express Request
+ * @param {object} req.params - Request params
  * @param {string} req.params.id - Assistant identifier.
  * @param {AssistantUpdateParams} req.body - The assistant update parameters.
  * @returns {Assistant} 200 - success response - application/json
  */
-router.patch('/:id', async (req, res) => {
+const patchAssistant = async (req, res) => {
   try {
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
+    await validateAuthor({ req, openai });
 
     const assistant_id = req.params.id;
-    const updateData = req.body;
+    const { endpoint: _e, ...updateData } = req.body;
     updateData.tools = (updateData.tools ?? [])
       .map((tool) => {
         if (typeof tool !== 'string') {
@@ -119,90 +109,76 @@ router.patch('/:id', async (req, res) => {
     logger.error('[/assistants/:id] Error updating assistant', error);
     res.status(500).json({ error: error.message });
   }
-});
+};
 
 /**
  * Deletes an assistant.
  * @route DELETE /assistants/:id
+ * @param {object} req - Express Request
+ * @param {object} req.params - Request params
  * @param {string} req.params.id - Assistant identifier.
  * @returns {Assistant} 200 - success response - application/json
  */
-router.delete('/:id', async (req, res) => {
+const deleteAssistant = async (req, res) => {
   try {
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
+    await validateAuthor({ req, openai });
 
     const assistant_id = req.params.id;
     const deletionStatus = await openai.beta.assistants.del(assistant_id);
+    if (deletionStatus?.deleted) {
+      await deleteAssistantActions({ req, assistant_id });
+    }
     res.json(deletionStatus);
   } catch (error) {
     logger.error('[/assistants/:id] Error deleting assistant', error);
     res.status(500).json({ error: 'Error deleting assistant' });
   }
-});
+};
 
 /**
  * Returns a list of assistants.
  * @route GET /assistants
+ * @param {object} req - Express Request
  * @param {AssistantListParams} req.query - The assistant list parameters for pagination and sorting.
  * @returns {AssistantListResponse} 200 - success response - application/json
  */
-router.get('/', async (req, res) => {
+const listAssistants = async (req, res) => {
   try {
-    const { limit = 100, order = 'desc', after, before } = req.query;
-    const query = { limit, order, after, before };
-
-    const azureConfig = req.app.locals[EModelEndpoint.azureOpenAI];
-    /** @type {AssistantListResponse} */
-    let body;
-
-    if (azureConfig?.assistants) {
-      body = await listAssistantsForAzure({ req, res, azureConfig, query });
-    } else {
-      ({ body } = await listAssistants({ req, res, query }));
-    }
-
-    if (req.app.locals?.[EModelEndpoint.assistants]) {
-      /** @type {Partial<TAssistantEndpoint>} */
-      const assistantsConfig = req.app.locals[EModelEndpoint.assistants];
-      const { supportedIds, excludedIds } = assistantsConfig;
-      if (supportedIds?.length) {
-        body.data = body.data.filter((assistant) => supportedIds.includes(assistant.id));
-      } else if (excludedIds?.length) {
-        body.data = body.data.filter((assistant) => !excludedIds.includes(assistant.id));
-      }
-    }
-
+    const body = await fetchAssistants({ req, res });
     res.json(body);
   } catch (error) {
     logger.error('[/assistants] Error listing assistants', error);
     res.status(500).json({ message: 'Error listing assistants' });
   }
-});
+};
 
 /**
  * Returns a list of the user's assistant documents (metadata saved to database).
  * @route GET /assistants/documents
  * @returns {AssistantDocument[]} 200 - success response - application/json
  */
-router.get('/documents', async (req, res) => {
+const getAssistantDocuments = async (req, res) => {
   try {
     res.json(await getAssistants({ user: req.user.id }));
   } catch (error) {
     logger.error('[/assistants/documents] Error listing assistant documents', error);
     res.status(500).json({ error: error.message });
   }
-});
+};
 
 /**
  * Uploads and updates an avatar for a specific assistant.
  * @route POST /avatar/:assistant_id
+ * @param {object} req - Express Request
+ * @param {object} req.params - Request params
  * @param {string} req.params.assistant_id - The ID of the assistant.
  * @param {Express.Multer.File} req.file - The avatar image file.
+ * @param {object} req.body - Request body
  * @param {string} [req.body.metadata] - Optional metadata for the assistant's avatar.
  * @returns {Object} 200 - success response - application/json
  */
-router.post('/avatar/:assistant_id', upload.single('file'), async (req, res) => {
+const uploadAssistantAvatar = async (req, res) => {
   try {
     const { assistant_id } = req.params;
     if (!assistant_id) {
@@ -210,8 +186,8 @@ router.post('/avatar/:assistant_id', upload.single('file'), async (req, res) =>
     }
 
     let { metadata: _metadata = '{}' } = req.body;
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
+    await validateAuthor({ req, openai });
 
     const image = await uploadImageBuffer({
       req,
@@ -246,7 +222,7 @@ router.post('/avatar/:assistant_id', upload.single('file'), async (req, res) =>
 
     const promises = [];
     promises.push(
-      updateAssistant(
+      updateAssistantDoc(
         { assistant_id },
         {
           avatar: {
@@ -266,6 +242,14 @@ router.post('/avatar/:assistant_id', upload.single('file'), async (req, res) =>
     logger.error(message, error);
     res.status(500).json({ message });
   }
-});
+};
 
-module.exports = router;
+module.exports = {
+  createAssistant,
+  retrieveAssistant,
+  patchAssistant,
+  deleteAssistant,
+  listAssistants,
+  getAssistantDocuments,
+  uploadAssistantAvatar,
+};

api/server/controllers/assistants/v2.js

@@ -0,0 +1,213 @@
+const { ToolCallTypes } = require('librechat-data-provider');
+const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
+const { validateAndUpdateTool } = require('~/server/services/ActionService');
+const { updateAssistantDoc } = require('~/models/Assistant');
+const { getOpenAIClient } = require('./helpers');
+const { logger } = require('~/config');
+
+/**
+ * Create an assistant.
+ * @route POST /assistants
+ * @param {AssistantCreateParams} req.body - The assistant creation parameters.
+ * @returns {Assistant} 201 - success response - application/json
+ */
+const createAssistant = async (req, res) => {
+  try {
+    /** @type {{ openai: OpenAIClient }} */
+    const { openai } = await getOpenAIClient({ req, res });
+
+    const { tools = [], endpoint, ...assistantData } = req.body;
+    assistantData.tools = tools
+      .map((tool) => {
+        if (typeof tool !== 'string') {
+          return tool;
+        }
+
+        return req.app.locals.availableTools[tool];
+      })
+      .filter((tool) => tool);
+
+    let azureModelIdentifier = null;
+    if (openai.locals?.azureOptions) {
+      azureModelIdentifier = assistantData.model;
+      assistantData.model = openai.locals.azureOptions.azureOpenAIApiDeploymentName;
+    }
+
+    assistantData.metadata = {
+      author: req.user.id,
+      endpoint,
+    };
+
+    const assistant = await openai.beta.assistants.create(assistantData);
+    const promise = updateAssistantDoc({ assistant_id: assistant.id }, { user: req.user.id });
+    if (azureModelIdentifier) {
+      assistant.model = azureModelIdentifier;
+    }
+    await promise;
+    logger.debug('/assistants/', assistant);
+    res.status(201).json(assistant);
+  } catch (error) {
+    logger.error('[/assistants] Error creating assistant', error);
+    res.status(500).json({ error: error.message });
+  }
+};
+
+/**
+ * Modifies an assistant.
+ * @param {object} params
+ * @param {Express.Request} params.req
+ * @param {OpenAIClient} params.openai
+ * @param {string} params.assistant_id
+ * @param {AssistantUpdateParams} params.updateData
+ * @returns {Promise<Assistant>} The updated assistant.
+ */
+const updateAssistant = async ({ req, openai, assistant_id, updateData }) => {
+  await validateAuthor({ req, openai });
+  const tools = [];
+
+  let hasFileSearch = false;
+  for (const tool of updateData.tools ?? []) {
+    let actualTool = typeof tool === 'string' ? req.app.locals.availableTools[tool] : tool;
+
+    if (!actualTool) {
+      continue;
+    }
+
+    if (actualTool.type === ToolCallTypes.FILE_SEARCH) {
+      hasFileSearch = true;
+    }
+
+    if (!actualTool.function) {
+      tools.push(actualTool);
+      continue;
+    }
+
+    const updatedTool = await validateAndUpdateTool({ req, tool: actualTool, assistant_id });
+    if (updatedTool) {
+      tools.push(updatedTool);
+    }
+  }
+
+  if (hasFileSearch && !updateData.tool_resources) {
+    const assistant = await openai.beta.assistants.retrieve(assistant_id);
+    updateData.tool_resources = assistant.tool_resources ?? null;
+  }
+
+  if (hasFileSearch && !updateData.tool_resources?.file_search) {
+    updateData.tool_resources = {
+      ...(updateData.tool_resources ?? {}),
+      file_search: {
+        vector_store_ids: [],
+      },
+    };
+  }
+
+  updateData.tools = tools;
+
+  if (openai.locals?.azureOptions && updateData.model) {
+    updateData.model = openai.locals.azureOptions.azureOpenAIApiDeploymentName;
+  }
+
+  return await openai.beta.assistants.update(assistant_id, updateData);
+};
+
+/**
+ * Modifies an assistant with the resource file id.
+ * @param {object} params
+ * @param {Express.Request} params.req
+ * @param {OpenAIClient} params.openai
+ * @param {string} params.assistant_id
+ * @param {string} params.tool_resource
+ * @param {string} params.file_id
+ * @param {AssistantUpdateParams} params.updateData
+ * @returns {Promise<Assistant>} The updated assistant.
+ */
+const addResourceFileId = async ({ req, openai, assistant_id, tool_resource, file_id }) => {
+  const assistant = await openai.beta.assistants.retrieve(assistant_id);
+  const { tool_resources = {} } = assistant;
+  if (tool_resources[tool_resource]) {
+    tool_resources[tool_resource].file_ids.push(file_id);
+  } else {
+    tool_resources[tool_resource] = { file_ids: [file_id] };
+  }
+
+  delete assistant.id;
+  return await updateAssistant({
+    req,
+    openai,
+    assistant_id,
+    updateData: { tools: assistant.tools, tool_resources },
+  });
+};
+
+/**
+ * Deletes a file ID from an assistant's resource.
+ * @param {object} params
+ * @param {Express.Request} params.req
+ * @param {OpenAIClient} params.openai
+ * @param {string} params.assistant_id
+ * @param {string} [params.tool_resource]
+ * @param {string} params.file_id
+ * @param {AssistantUpdateParams} params.updateData
+ * @returns {Promise<Assistant>} The updated assistant.
+ */
+const deleteResourceFileId = async ({ req, openai, assistant_id, tool_resource, file_id }) => {
+  const assistant = await openai.beta.assistants.retrieve(assistant_id);
+  const { tool_resources = {} } = assistant;
+
+  if (tool_resource && tool_resources[tool_resource]) {
+    const resource = tool_resources[tool_resource];
+    const index = resource.file_ids.indexOf(file_id);
+    if (index !== -1) {
+      resource.file_ids.splice(index, 1);
+    }
+  } else {
+    for (const resourceKey in tool_resources) {
+      const resource = tool_resources[resourceKey];
+      const index = resource.file_ids.indexOf(file_id);
+      if (index !== -1) {
+        resource.file_ids.splice(index, 1);
+        break;
+      }
+    }
+  }
+
+  delete assistant.id;
+  return await updateAssistant({
+    req,
+    openai,
+    assistant_id,
+    updateData: { tools: assistant.tools, tool_resources },
+  });
+};
+
+/**
+ * Modifies an assistant.
+ * @route PATCH /assistants/:id
+ * @param {object} req - Express Request
+ * @param {object} req.params - Request params
+ * @param {string} req.params.id - Assistant identifier.
+ * @param {AssistantUpdateParams} req.body - The assistant update parameters.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+const patchAssistant = async (req, res) => {
+  try {
+    const { openai } = await getOpenAIClient({ req, res });
+    const assistant_id = req.params.id;
+    const { endpoint: _e, ...updateData } = req.body;
+    updateData.tools = updateData.tools ?? [];
+    const updatedAssistant = await updateAssistant({ req, openai, assistant_id, updateData });
+    res.json(updatedAssistant);
+  } catch (error) {
+    logger.error('[/assistants/:id] Error updating assistant', error);
+    res.status(500).json({ error: error.message });
+  }
+};
+
+module.exports = {
+  patchAssistant,
+  createAssistant,
+  updateAssistant,
+  addResourceFileId,
+  deleteResourceFileId,
+};

api/server/controllers/auth/LoginController.js

@@ -1,26 +1,22 @@
-const User = require('~/models/User');
 const { setAuthTokens } = require('~/server/services/AuthService');
 const { logger } = require('~/config');
 
 const loginController = async (req, res) => {
   try {
-    const user = await User.findById(req.user._id);
-
-    // If user doesn't exist, return error
-    if (!user) {
-      // typeof user !== User) { // this doesn't seem to resolve the User type ??
+    if (!req.user) {
       return res.status(400).json({ message: 'Invalid credentials' });
     }
 
-    const token = await setAuthTokens(user._id, res);
+    const { password: _, __v, ...user } = req.user;
+    user.id = user._id.toString();
+
+    const token = await setAuthTokens(req.user._id, res);
 
     return res.status(200).send({ token, user });
   } catch (err) {
     logger.error('[loginController]', err);
+    return res.status(500).json({ message: 'Something went wrong' });
   }
-
-  // Generic error messages are safer
-  return res.status(500).json({ message: 'Something went wrong' });
 };
 
 module.exports = {

api/server/index.js

@@ -6,16 +6,16 @@ const axios = require('axios');
 const express = require('express');
 const passport = require('passport');
 const mongoSanitize = require('express-mongo-sanitize');
+const { jwtLogin, passportLogin } = require('~/strategies');
+const { connectDb, indexSync } = require('~/lib/db');
+const { isEnabled } = require('~/server/utils');
+const { ldapLogin } = require('~/strategies');
+const { logger } = require('~/config');
 const validateImageRequest = require('./middleware/validateImageRequest');
 const errorController = require('./controllers/ErrorController');
-const { jwtLogin, passportLogin } = require('~/strategies');
 const configureSocialLogins = require('./socialLogins');
-const { connectDb, indexSync } = require('~/lib/db');
 const AppService = require('./services/AppService');
 const noIndex = require('./middleware/noIndex');
-const { isEnabled } = require('~/server/utils');
-const { logger } = require('~/config');
-
 const routes = require('./routes');
 
 const { PORT, HOST, ALLOW_SOCIAL_LOGIN } = process.env ?? {};
@@ -60,6 +60,11 @@ const startServer = async () => {
   passport.use(await jwtLogin());
   passport.use(passportLogin());
 
+  // LDAP Auth
+  if (process.env.LDAP_URL && process.env.LDAP_BIND_DN && process.env.LDAP_USER_SEARCH_BASE) {
+    passport.use(ldapLogin);
+  }
+
   if (isEnabled(ALLOW_SOCIAL_LOGIN)) {
     configureSocialLogins(app);
   }
@@ -85,9 +90,10 @@ const startServer = async () => {
   app.use('/api/assistants', routes.assistants);
   app.use('/api/files', await routes.files.initialize());
   app.use('/images/', validateImageRequest, routes.staticRoute);
+  app.use('/api/share', routes.share);
 
   app.use((req, res) => {
-    res.status(404).sendFile(path.join(app.locals.paths.dist, 'index.html'));
+    res.sendFile(path.join(app.locals.paths.dist, 'index.html'));
   });
 
   app.listen(port, host, () => {

api/server/middleware/abortMiddleware.js

@@ -1,4 +1,4 @@
-const { EModelEndpoint } = require('librechat-data-provider');
+const { isAssistantsEndpoint } = require('librechat-data-provider');
 const { sendMessage, sendError, countTokens, isEnabled } = require('~/server/utils');
 const { truncateText, smartTruncateText } = require('~/app/clients/prompts');
 const { saveMessage, getConvo, getConvoTitle } = require('~/models');
@@ -15,7 +15,7 @@ async function abortMessage(req, res) {
     abortKey = conversationId;
   }
 
-  if (endpoint === EModelEndpoint.assistants) {
+  if (isAssistantsEndpoint(endpoint)) {
     return await abortRun(req, res);
   }
 

api/server/middleware/abortRun.js

@@ -1,6 +1,7 @@
 const { CacheKeys, RunStatus, isUUID } = require('librechat-data-provider');
 const { initializeClient } = require('~/server/services/Endpoints/assistants');
 const { checkMessageGaps, recordUsage } = require('~/server/services/Threads');
+const { deleteMessages } = require('~/models/Message');
 const { getConvo } = require('~/models/Conversation');
 const getLogStores = require('~/cache/getLogStores');
 const { sendMessage } = require('~/server/utils');
@@ -10,7 +11,7 @@ const three_minutes = 1000 * 60 * 3;
 
 async function abortRun(req, res) {
   res.setHeader('Content-Type', 'application/json');
-  const { abortKey } = req.body;
+  const { abortKey, endpoint } = req.body;
   const [conversationId, latestMessageId] = abortKey.split(':');
   const conversation = await getConvo(req.user.id, conversationId);
 
@@ -66,12 +67,19 @@ async function abortRun(req, res) {
     logger.error('[abortRun] Error fetching or processing run', error);
   }
 
+  /* TODO: a reconciling strategy between the existing intermediate message would be more optimal than deleting it */
+  await deleteMessages({
+    user: req.user.id,
+    unfinished: true,
+    conversationId,
+  });
   runMessages = await checkMessageGaps({
     openai,
-    latestMessageId,
-    thread_id,
     run_id,
+    endpoint,
+    thread_id,
     conversationId,
+    latestMessageId,
   });
 
   const finalEvent = {

api/server/middleware/assistants/validate.js

@@ -0,0 +1,43 @@
+const { v4 } = require('uuid');
+const { handleAbortError } = require('~/server/middleware/abortMiddleware');
+
+/**
+ * Checks if the assistant is supported or excluded
+ * @param {object} req - Express Request
+ * @param {object} req.body - The request payload.
+ * @param {object} res - Express Response
+ * @param {function} next - Express next middleware function.
+ * @returns {Promise<void>}
+ */
+const validateAssistant = async (req, res, next) => {
+  const { endpoint, conversationId, assistant_id, messageId } = req.body;
+
+  /** @type {Partial<TAssistantEndpoint>} */
+  const assistantsConfig = req.app.locals?.[endpoint];
+  if (!assistantsConfig) {
+    return next();
+  }
+
+  const { supportedIds, excludedIds } = assistantsConfig;
+  const error = { message: 'Assistant not supported' };
+  if (supportedIds?.length && !supportedIds.includes(assistant_id)) {
+    return await handleAbortError(res, req, error, {
+      sender: 'System',
+      conversationId,
+      messageId: v4(),
+      parentMessageId: messageId,
+      error,
+    });
+  } else if (excludedIds?.length && excludedIds.includes(assistant_id)) {
+    return await handleAbortError(res, req, error, {
+      sender: 'System',
+      conversationId,
+      messageId: v4(),
+      parentMessageId: messageId,
+    });
+  }
+
+  return next();
+};
+
+module.exports = validateAssistant;

api/server/middleware/assistants/validateAuthor.js

@@ -0,0 +1,42 @@
+const { getAssistant } = require('~/models/Assistant');
+
+/**
+ * Checks if the assistant is supported or excluded
+ * @param {object} params
+ * @param {object} params.req - Express Request
+ * @param {object} params.req.body - The request payload.
+ * @param {string} params.overrideEndpoint - The override endpoint
+ * @param {string} params.overrideAssistantId - The override assistant ID
+ * @param {OpenAIClient} params.openai - OpenAI API Client
+ * @returns {Promise<void>}
+ */
+const validateAuthor = async ({ req, openai, overrideEndpoint, overrideAssistantId }) => {
+  if (req.user.role === 'ADMIN') {
+    return;
+  }
+
+  const endpoint = overrideEndpoint ?? req.body.endpoint ?? req.query.endpoint;
+  const assistant_id =
+    overrideAssistantId ?? req.params.id ?? req.body.assistant_id ?? req.query.assistant_id;
+
+  /** @type {Partial<TAssistantEndpoint>} */
+  const assistantsConfig = req.app.locals?.[endpoint];
+  if (!assistantsConfig) {
+    return;
+  }
+
+  if (!assistantsConfig.privateAssistants) {
+    return;
+  }
+
+  const assistantDoc = await getAssistant({ assistant_id, user: req.user.id });
+  if (assistantDoc) {
+    return;
+  }
+  const assistant = await openai.beta.assistants.retrieve(assistant_id);
+  if (req.user.id !== assistant?.metadata?.author) {
+    throw new Error(`Assistant ${assistant_id} is not authored by the user.`);
+  }
+};
+
+module.exports = validateAuthor;

api/server/middleware/buildEndpointOption.js

@@ -1,5 +1,6 @@
 const { parseConvo, EModelEndpoint } = require('librechat-data-provider');
 const { getModelsConfig } = require('~/server/controllers/ModelController');
+const azureAssistants = require('~/server/services/Endpoints/azureAssistants');
 const assistants = require('~/server/services/Endpoints/assistants');
 const gptPlugins = require('~/server/services/Endpoints/gptPlugins');
 const { processFiles } = require('~/server/services/Files/process');
@@ -18,6 +19,7 @@ const buildFunction = {
   [EModelEndpoint.anthropic]: anthropic.buildOptions,
   [EModelEndpoint.gptPlugins]: gptPlugins.buildOptions,
   [EModelEndpoint.assistants]: assistants.buildOptions,
+  [EModelEndpoint.azureAssistants]: azureAssistants.buildOptions,
 };
 
 async function buildEndpointOption(req, res, next) {
@@ -42,6 +44,15 @@ async function buildEndpointOption(req, res, next) {
       return handleError(res, { text: 'Model spec mismatch' });
     }
 
+    if (
+      currentModelSpec.preset.endpoint !== EModelEndpoint.gptPlugins &&
+      currentModelSpec.preset.tools
+    ) {
+      return handleError(res, {
+        text: `Only the "${EModelEndpoint.gptPlugins}" endpoint can have tools defined in the preset`,
+      });
+    }
+
     const isValidModelSpec = enforceModelSpec(currentModelSpec, parsedBody);
     if (!isValidModelSpec) {
       return handleError(res, { text: 'Model spec mismatch' });

api/server/middleware/canDeleteAccount.js

@@ -0,0 +1,27 @@
+const { isEnabled } = require('~/server/utils');
+const { logger } = require('~/config');
+
+/**
+ * Checks if the user can delete their account
+ *
+ * @async
+ * @function
+ * @param {Object} req - Express request object
+ * @param {Object} res - Express response object
+ * @param {Function} next - Next middleware function
+ *
+ * @returns {Promise<function|Object>} - Returns a Promise which when resolved calls next middleware if the user can delete their account
+ */
+
+const canDeleteAccount = async (req, res, next = () => {}) => {
+  const { user } = req;
+  const { ALLOW_ACCOUNT_DELETION = true } = process.env;
+  if (user?.role === 'ADMIN' || isEnabled(ALLOW_ACCOUNT_DELETION)) {
+    return next();
+  } else {
+    logger.error(`[User] [Delete Account] [User cannot delete account] [User: ${user?.id}]`);
+    return res.status(403).send({ message: 'You do not have permission to delete this account' });
+  }
+};
+
+module.exports = canDeleteAccount;

api/server/middleware/checkBan.js

@@ -1,15 +1,13 @@
 const Keyv = require('keyv');
 const uap = require('ua-parser-js');
 const { ViolationTypes } = require('librechat-data-provider');
-const { isEnabled, removePorts } = require('../utils');
-const keyvRedis = require('~/cache/keyvRedis');
+const { isEnabled, removePorts } = require('~/server/utils');
+const keyvMongo = require('~/cache/keyvMongo');
 const denyRequest = require('./denyRequest');
 const { getLogStores } = require('~/cache');
-const User = require('~/models/User');
+const { findUser } = require('~/models');
 
-const banCache = isEnabled(process.env.USE_REDIS)
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: ViolationTypes.BAN, ttl: 0 });
+const banCache = new Keyv({ store: keyvMongo, namespace: ViolationTypes.BAN, ttl: 0 });
 const message = 'Your account has been temporarily banned due to violations of our service.';
 
 /**
@@ -57,7 +55,7 @@ const checkBan = async (req, res, next = () => {}) => {
   let userId = req.user?.id ?? req.user?._id ?? null;
 
   if (!userId && req?.body?.email) {
-    const user = await User.findOne({ email: req.body.email }, '_id').lean();
+    const user = await findUser({ email: req.body.email }, '_id');
     userId = user?._id ? user._id.toString() : userId;
   }
 

api/server/middleware/enforceModelSpec.js

@@ -24,21 +24,32 @@ const enforceModelSpec = (modelSpec, parsedBody) => {
 
 /**
  * Checks if there is a match for the given key and value in the parsed body
- * or any of its interchangeable keys.
+ * or any of its interchangeable keys, including deep comparison for objects and arrays.
  * @param {string} key
  * @param {any} value
- * @param {TConversation} parsedBody
+ * @param {object} parsedBody
  * @returns {boolean}
  */
 const checkMatch = (key, value, parsedBody) => {
-  if (parsedBody[key] === value) {
+  const isEqual = (a, b) => {
+    if (Array.isArray(a) && Array.isArray(b)) {
+      return a.length === b.length && a.every((val, index) => isEqual(val, b[index]));
+    } else if (typeof a === 'object' && typeof b === 'object' && a !== null && b !== null) {
+      const keysA = Object.keys(a);
+      const keysB = Object.keys(b);
+      return keysA.length === keysB.length && keysA.every((k) => isEqual(a[k], b[k]));
+    }
+    return a === b;
+  };
+
+  if (isEqual(parsedBody[key], value)) {
     return true;
   }
 
   if (interchangeableKeys.has(key)) {
     return interchangeableKeys
       .get(key)
-      .some((interchangeableKey) => parsedBody[interchangeableKey] === value);
+      .some((interchangeableKey) => isEqual(parsedBody[interchangeableKey], value));
   }
 
   return false;

api/server/middleware/enforceModelSpec.spec.js

@@ -0,0 +1,47 @@
+// enforceModelSpec.test.js
+
+const enforceModelSpec = require('./enforceModelSpec');
+
+describe('enforceModelSpec function', () => {
+  test('returns true when all model specs match parsed body directly', () => {
+    const modelSpec = { preset: { title: 'Dialog', status: 'Active' } };
+    const parsedBody = { title: 'Dialog', status: 'Active' };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(true);
+  });
+
+  test('returns true when model specs match via interchangeable keys', () => {
+    const modelSpec = { preset: { chatGptLabel: 'GPT-4' } };
+    const parsedBody = { modelLabel: 'GPT-4' };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(true);
+  });
+
+  test('returns false if any key value does not match', () => {
+    const modelSpec = { preset: { language: 'English', level: 'Advanced' } };
+    const parsedBody = { language: 'Spanish', level: 'Advanced' };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(false);
+  });
+
+  test('ignores the \'endpoint\' key in model spec', () => {
+    const modelSpec = { preset: { endpoint: 'ignored', feature: 'Special' } };
+    const parsedBody = { feature: 'Special' };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(true);
+  });
+
+  test('handles nested objects correctly', () => {
+    const modelSpec = { preset: { details: { time: 'noon', location: 'park' } } };
+    const parsedBody = { details: { time: 'noon', location: 'park' } };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(true);
+  });
+
+  test('handles arrays within objects', () => {
+    const modelSpec = { preset: { tags: ['urgent', 'important'] } };
+    const parsedBody = { tags: ['urgent', 'important'] };
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(true);
+  });
+
+  test('fails when arrays in objects do not match', () => {
+    const modelSpec = { preset: { tags: ['urgent', 'important'] } };
+    const parsedBody = { tags: ['important', 'urgent'] }; // Different order
+    expect(enforceModelSpec(modelSpec, parsedBody)).toBe(false);
+  });
+});

api/server/middleware/index.js

@@ -1,45 +1,43 @@
-const abortMiddleware = require('./abortMiddleware');
-const checkBan = require('./checkBan');
-const checkDomainAllowed = require('./checkDomainAllowed');
-const uaParser = require('./uaParser');
-const setHeaders = require('./setHeaders');
-const loginLimiter = require('./loginLimiter');
-const validateModel = require('./validateModel');
-const requireJwtAuth = require('./requireJwtAuth');
-const uploadLimiters = require('./uploadLimiters');
-const registerLimiter = require('./registerLimiter');
-const messageLimiters = require('./messageLimiters');
-const requireLocalAuth = require('./requireLocalAuth');
-const validateEndpoint = require('./validateEndpoint');
-const concurrentLimiter = require('./concurrentLimiter');
-const validateMessageReq = require('./validateMessageReq');
-const buildEndpointOption = require('./buildEndpointOption');
+const validatePasswordReset = require('./validatePasswordReset');
 const validateRegistration = require('./validateRegistration');
 const validateImageRequest = require('./validateImageRequest');
+const buildEndpointOption = require('./buildEndpointOption');
+const validateMessageReq = require('./validateMessageReq');
+const checkDomainAllowed = require('./checkDomainAllowed');
+const concurrentLimiter = require('./concurrentLimiter');
+const validateEndpoint = require('./validateEndpoint');
+const requireLocalAuth = require('./requireLocalAuth');
+const canDeleteAccount = require('./canDeleteAccount');
+const requireLdapAuth = require('./requireLdapAuth');
+const abortMiddleware = require('./abortMiddleware');
+const requireJwtAuth = require('./requireJwtAuth');
+const validateModel = require('./validateModel');
 const moderateText = require('./moderateText');
+const setHeaders = require('./setHeaders');
+const limiters = require('./limiters');
+const uaParser = require('./uaParser');
+const checkBan = require('./checkBan');
 const noIndex = require('./noIndex');
-const importLimiters = require('./importLimiters');
 
 module.exports = {
-  ...uploadLimiters,
   ...abortMiddleware,
-  ...messageLimiters,
+  ...limiters,
+  noIndex,
   checkBan,
   uaParser,
   setHeaders,
-  loginLimiter,
+  moderateText,
+  validateModel,
   requireJwtAuth,
-  registerLimiter,
+  requireLdapAuth,
   requireLocalAuth,
+  canDeleteAccount,
   validateEndpoint,
   concurrentLimiter,
+  checkDomainAllowed,
   validateMessageReq,
   buildEndpointOption,
   validateRegistration,
   validateImageRequest,
-  validateModel,
-  moderateText,
-  noIndex,
-  ...importLimiters,
-  checkDomainAllowed,
+  validatePasswordReset,
 };

api/server/middleware/limiters/index.js

@@ -0,0 +1,22 @@
+const createTTSLimiters = require('./ttsLimiters');
+const createSTTLimiters = require('./sttLimiters');
+
+const loginLimiter = require('./loginLimiter');
+const importLimiters = require('./importLimiters');
+const uploadLimiters = require('./uploadLimiters');
+const registerLimiter = require('./registerLimiter');
+const messageLimiters = require('./messageLimiters');
+const verifyEmailLimiter = require('./verifyEmailLimiter');
+const resetPasswordLimiter = require('./resetPasswordLimiter');
+
+module.exports = {
+  ...uploadLimiters,
+  ...importLimiters,
+  ...messageLimiters,
+  loginLimiter,
+  registerLimiter,
+  createTTSLimiters,
+  createSTTLimiters,
+  verifyEmailLimiter,
+  resetPasswordLimiter,
+};

api/server/middleware/limiters/loginLimiter.js

@@ -1,6 +1,6 @@
 const rateLimit = require('express-rate-limit');
-const { logViolation } = require('../../cache');
-const { removePorts } = require('../utils');
+const { removePorts } = require('~/server/utils');
+const { logViolation } = require('~/cache');
 
 const { LOGIN_WINDOW = 5, LOGIN_MAX = 7, LOGIN_VIOLATION_SCORE: score } = process.env;
 const windowMs = LOGIN_WINDOW * 60 * 1000;

api/server/middleware/limiters/messageLimiters.js

@@ -1,6 +1,6 @@
 const rateLimit = require('express-rate-limit');
-const { logViolation } = require('../../cache');
-const denyRequest = require('./denyRequest');
+const denyRequest = require('~/server/middleware/denyRequest');
+const { logViolation } = require('~/cache');
 
 const {
   MESSAGE_IP_MAX = 40,

api/server/middleware/limiters/registerLimiter.js

@@ -1,6 +1,6 @@
 const rateLimit = require('express-rate-limit');
-const { logViolation } = require('../../cache');
-const { removePorts } = require('../utils');
+const { removePorts } = require('~/server/utils');
+const { logViolation } = require('~/cache');
 
 const { REGISTER_WINDOW = 60, REGISTER_MAX = 5, REGISTRATION_VIOLATION_SCORE: score } = process.env;
 const windowMs = REGISTER_WINDOW * 60 * 1000;

api/server/middleware/limiters/resetPasswordLimiter.js

@@ -0,0 +1,35 @@
+const rateLimit = require('express-rate-limit');
+const { ViolationTypes } = require('librechat-data-provider');
+const { removePorts } = require('~/server/utils');
+const { logViolation } = require('~/cache');
+
+const {
+  RESET_PASSWORD_WINDOW = 2,
+  RESET_PASSWORD_MAX = 2,
+  RESET_PASSWORD_VIOLATION_SCORE: score,
+} = process.env;
+const windowMs = RESET_PASSWORD_WINDOW * 60 * 1000;
+const max = RESET_PASSWORD_MAX;
+const windowInMinutes = windowMs / 60000;
+const message = `Too many attempts, please try again after ${windowInMinutes} minute(s)`;
+
+const handler = async (req, res) => {
+  const type = ViolationTypes.RESET_PASSWORD_LIMIT;
+  const errorMessage = {
+    type,
+    max,
+    windowInMinutes,
+  };
+
+  await logViolation(req, res, type, errorMessage, score);
+  return res.status(429).json({ message });
+};
+
+const resetPasswordLimiter = rateLimit({
+  windowMs,
+  max,
+  handler,
+  keyGenerator: removePorts,
+});
+
+module.exports = resetPasswordLimiter;

api/server/middleware/limiters/sttLimiters.js

@@ -0,0 +1,68 @@
+const rateLimit = require('express-rate-limit');
+const { ViolationTypes } = require('librechat-data-provider');
+const logViolation = require('~/cache/logViolation');
+
+const getEnvironmentVariables = () => {
+  const STT_IP_MAX = parseInt(process.env.STT_IP_MAX) || 100;
+  const STT_IP_WINDOW = parseInt(process.env.STT_IP_WINDOW) || 1;
+  const STT_USER_MAX = parseInt(process.env.STT_USER_MAX) || 50;
+  const STT_USER_WINDOW = parseInt(process.env.STT_USER_WINDOW) || 1;
+
+  const sttIpWindowMs = STT_IP_WINDOW * 60 * 1000;
+  const sttIpMax = STT_IP_MAX;
+  const sttIpWindowInMinutes = sttIpWindowMs / 60000;
+
+  const sttUserWindowMs = STT_USER_WINDOW * 60 * 1000;
+  const sttUserMax = STT_USER_MAX;
+  const sttUserWindowInMinutes = sttUserWindowMs / 60000;
+
+  return {
+    sttIpWindowMs,
+    sttIpMax,
+    sttIpWindowInMinutes,
+    sttUserWindowMs,
+    sttUserMax,
+    sttUserWindowInMinutes,
+  };
+};
+
+const createSTTHandler = (ip = true) => {
+  const { sttIpMax, sttIpWindowInMinutes, sttUserMax, sttUserWindowInMinutes } =
+    getEnvironmentVariables();
+
+  return async (req, res) => {
+    const type = ViolationTypes.STT_LIMIT;
+    const errorMessage = {
+      type,
+      max: ip ? sttIpMax : sttUserMax,
+      limiter: ip ? 'ip' : 'user',
+      windowInMinutes: ip ? sttIpWindowInMinutes : sttUserWindowInMinutes,
+    };
+
+    await logViolation(req, res, type, errorMessage);
+    res.status(429).json({ message: 'Too many STT requests. Try again later' });
+  };
+};
+
+const createSTTLimiters = () => {
+  const { sttIpWindowMs, sttIpMax, sttUserWindowMs, sttUserMax } = getEnvironmentVariables();
+
+  const sttIpLimiter = rateLimit({
+    windowMs: sttIpWindowMs,
+    max: sttIpMax,
+    handler: createSTTHandler(),
+  });
+
+  const sttUserLimiter = rateLimit({
+    windowMs: sttUserWindowMs,
+    max: sttUserMax,
+    handler: createSTTHandler(false),
+    keyGenerator: function (req) {
+      return req.user?.id; // Use the user ID or NULL if not available
+    },
+  });
+
+  return { sttIpLimiter, sttUserLimiter };
+};
+
+module.exports = createSTTLimiters;

api/server/middleware/limiters/ttsLimiters.js

@@ -0,0 +1,68 @@
+const rateLimit = require('express-rate-limit');
+const { ViolationTypes } = require('librechat-data-provider');
+const logViolation = require('~/cache/logViolation');
+
+const getEnvironmentVariables = () => {
+  const TTS_IP_MAX = parseInt(process.env.TTS_IP_MAX) || 100;
+  const TTS_IP_WINDOW = parseInt(process.env.TTS_IP_WINDOW) || 1;
+  const TTS_USER_MAX = parseInt(process.env.TTS_USER_MAX) || 50;
+  const TTS_USER_WINDOW = parseInt(process.env.TTS_USER_WINDOW) || 1;
+
+  const ttsIpWindowMs = TTS_IP_WINDOW * 60 * 1000;
+  const ttsIpMax = TTS_IP_MAX;
+  const ttsIpWindowInMinutes = ttsIpWindowMs / 60000;
+
+  const ttsUserWindowMs = TTS_USER_WINDOW * 60 * 1000;
+  const ttsUserMax = TTS_USER_MAX;
+  const ttsUserWindowInMinutes = ttsUserWindowMs / 60000;
+
+  return {
+    ttsIpWindowMs,
+    ttsIpMax,
+    ttsIpWindowInMinutes,
+    ttsUserWindowMs,
+    ttsUserMax,
+    ttsUserWindowInMinutes,
+  };
+};
+
+const createTTSHandler = (ip = true) => {
+  const { ttsIpMax, ttsIpWindowInMinutes, ttsUserMax, ttsUserWindowInMinutes } =
+    getEnvironmentVariables();
+
+  return async (req, res) => {
+    const type = ViolationTypes.TTS_LIMIT;
+    const errorMessage = {
+      type,
+      max: ip ? ttsIpMax : ttsUserMax,
+      limiter: ip ? 'ip' : 'user',
+      windowInMinutes: ip ? ttsIpWindowInMinutes : ttsUserWindowInMinutes,
+    };
+
+    await logViolation(req, res, type, errorMessage);
+    res.status(429).json({ message: 'Too many TTS requests. Try again later' });
+  };
+};
+
+const createTTSLimiters = () => {
+  const { ttsIpWindowMs, ttsIpMax, ttsUserWindowMs, ttsUserMax } = getEnvironmentVariables();
+
+  const ttsIpLimiter = rateLimit({
+    windowMs: ttsIpWindowMs,
+    max: ttsIpMax,
+    handler: createTTSHandler(),
+  });
+
+  const ttsUserLimiter = rateLimit({
+    windowMs: ttsUserWindowMs,
+    max: ttsUserMax,
+    handler: createTTSHandler(false),
+    keyGenerator: function (req) {
+      return req.user?.id; // Use the user ID or NULL if not available
+    },
+  });
+
+  return { ttsIpLimiter, ttsUserLimiter };
+};
+
+module.exports = createTTSLimiters;

api/server/middleware/limiters/verifyEmailLimiter.js

@@ -0,0 +1,35 @@
+const rateLimit = require('express-rate-limit');
+const { ViolationTypes } = require('librechat-data-provider');
+const { removePorts } = require('~/server/utils');
+const { logViolation } = require('~/cache');
+
+const {
+  VERIFY_EMAIL_WINDOW = 2,
+  VERIFY_EMAIL_MAX = 2,
+  VERIFY_EMAIL_VIOLATION_SCORE: score,
+} = process.env;
+const windowMs = VERIFY_EMAIL_WINDOW * 60 * 1000;
+const max = VERIFY_EMAIL_MAX;
+const windowInMinutes = windowMs / 60000;
+const message = `Too many attempts, please try again after ${windowInMinutes} minute(s)`;
+
+const handler = async (req, res) => {
+  const type = ViolationTypes.VERIFY_EMAIL_LIMIT;
+  const errorMessage = {
+    type,
+    max,
+    windowInMinutes,
+  };
+
+  await logViolation(req, res, type, errorMessage, score);
+  return res.status(429).json({ message });
+};
+
+const verifyEmailLimiter = rateLimit({
+  windowMs,
+  max,
+  handler,
+  keyGenerator: removePorts,
+});
+
+module.exports = verifyEmailLimiter;

api/server/middleware/requireLdapAuth.js

@@ -0,0 +1,22 @@
+const passport = require('passport');
+
+const requireLdapAuth = (req, res, next) => {
+  passport.authenticate('ldapauth', (err, user, info) => {
+    if (err) {
+      console.log({
+        title: '(requireLdapAuth) Error at passport.authenticate',
+        parameters: [{ name: 'error', value: err }],
+      });
+      return next(err);
+    }
+    if (!user) {
+      console.log({
+        title: '(requireLdapAuth) Error: No user',
+      });
+      return res.status(422).send(info);
+    }
+    req.user = user;
+    next();
+  })(req, res, next);
+};
+module.exports = requireLdapAuth;

api/server/middleware/requireLocalAuth.js

@@ -21,7 +21,13 @@ const requireLocalAuth = (req, res, next) => {
       log({
         title: '(requireLocalAuth) Error: No user',
       });
-      return res.status(422).send(info);
+      return res.status(404).send(info);
+    }
+    if (info && info.message) {
+      log({
+        title: '(requireLocalAuth) Error: ' + info.message,
+      });
+      return res.status(422).send({ message: info.message });
     }
     req.user = user;
     next();

api/server/middleware/validatePasswordReset.js

@@ -0,0 +1,13 @@
+const { isEnabled } = require('~/server/utils');
+const { logger } = require('~/config');
+
+function validatePasswordReset(req, res, next) {
+  if (isEnabled(process.env.ALLOW_PASSWORD_RESET)) {
+    next();
+  } else {
+    logger.warn(`Password reset attempt while not allowed. IP: ${req.ip}`);
+    res.status(403).send('Password reset is not allowed.');
+  }
+}
+
+module.exports = validatePasswordReset;

api/server/middleware/validateRegistration.js

@@ -1,6 +1,7 @@
+const { isEnabled } = require('~/server/utils');
+
 function validateRegistration(req, res, next) {
-  const setting = process.env.ALLOW_REGISTRATION?.toLowerCase();
-  if (setting === 'true') {
+  if (isEnabled(process.env.ALLOW_REGISTRATION)) {
     next();
   } else {
     res.status(403).send('Registration is not allowed.');

api/server/routes/__tests__/config.spec.js

@@ -25,6 +25,12 @@ afterEach(() => {
   delete process.env.DOMAIN_SERVER;
   delete process.env.ALLOW_REGISTRATION;
   delete process.env.ALLOW_SOCIAL_LOGIN;
+  delete process.env.ALLOW_PASSWORD_RESET;
+  delete process.env.LDAP_URL;
+  delete process.env.LDAP_BIND_DN;
+  delete process.env.LDAP_BIND_CREDENTIALS;
+  delete process.env.LDAP_USER_SEARCH_BASE;
+  delete process.env.LDAP_SEARCH_FILTER;
 });
 
 //TODO: This works/passes locally but http request tests fail with 404 in CI. Need to figure out why.
@@ -50,6 +56,12 @@ describe.skip('GET /', () => {
     process.env.DOMAIN_SERVER = 'http://test-server.com';
     process.env.ALLOW_REGISTRATION = 'true';
     process.env.ALLOW_SOCIAL_LOGIN = 'true';
+    process.env.ALLOW_PASSWORD_RESET = 'true';
+    process.env.LDAP_URL = 'Test LDAP URL';
+    process.env.LDAP_BIND_DN = 'Test LDAP Bind DN';
+    process.env.LDAP_BIND_CREDENTIALS = 'Test LDAP Bind Credentials';
+    process.env.LDAP_USER_SEARCH_BASE = 'Test LDAP User Search Base';
+    process.env.LDAP_SEARCH_FILTER = 'Test LDAP Search Filter';
 
     const response = await request(app).get('/');
 
@@ -64,9 +76,11 @@ describe.skip('GET /', () => {
       openidLoginEnabled: true,
       openidLabel: 'Test OpenID',
       openidImageUrl: 'http://test-server.com',
+      ldapLoginEnabled: true,
       serverDomain: 'http://test-server.com',
       emailLoginEnabled: 'true',
       registrationEnabled: 'true',
+      passwordResetEnabled: 'true',
       socialLoginEnabled: 'true',
     });
   });

api/server/routes/ask/gptPlugins.js

@@ -106,7 +106,11 @@ router.post(
     const pluginMap = new Map();
     const onAgentAction = async (action, runId) => {
       pluginMap.set(runId, action.tool);
-      sendIntermediateMessage(res, { plugins });
+      sendIntermediateMessage(res, {
+        plugins,
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
     };
 
     const onToolStart = async (tool, input, runId, parentRunId) => {
@@ -124,7 +128,11 @@ router.post(
       }
       const extraTokens = ':::plugin:::\n';
       plugins.push(latestPlugin);
-      sendIntermediateMessage(res, { plugins }, extraTokens);
+      sendIntermediateMessage(
+        res,
+        { plugins, parentMessageId: userMessage.messageId, messageId: responseMessageId },
+        extraTokens,
+      );
     };
 
     const onToolEnd = async (output, runId) => {
@@ -142,7 +150,11 @@ router.post(
 
     const onChainEnd = () => {
       saveMessage({ ...userMessage, user });
-      sendIntermediateMessage(res, { plugins });
+      sendIntermediateMessage(res, {
+        plugins,
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
     };
 
     const getAbortData = () => ({
@@ -174,12 +186,13 @@ router.post(
         onStart,
         getPartialText,
         ...endpointOption,
-        onProgress: progressCallback.call(null, {
+        progressCallback,
+        progressOptions: {
           res,
           text,
-          parentMessageId: overrideParentMessageId || userMessageId,
+          // parentMessageId: overrideParentMessageId || userMessageId,
           plugins,
-        }),
+        },
         abortController,
       });
 

api/server/routes/assistants/actions.js

@@ -2,9 +2,9 @@ const { v4 } = require('uuid');
 const express = require('express');
 const { encryptMetadata, domainParser } = require('~/server/services/ActionService');
 const { actionDelimiter, EModelEndpoint } = require('librechat-data-provider');
-const { initializeClient } = require('~/server/services/Endpoints/assistants');
+const { getOpenAIClient } = require('~/server/controllers/assistants/helpers');
 const { updateAction, getActions, deleteAction } = require('~/models/Action');
-const { updateAssistant, getAssistant } = require('~/models/Assistant');
+const { updateAssistantDoc, getAssistant } = require('~/models/Assistant');
 const { logger } = require('~/config');
 
 const router = express.Router();
@@ -45,7 +45,6 @@ router.post('/:assistant_id', async (req, res) => {
     let metadata = encryptMetadata(_metadata);
 
     let { domain } = metadata;
-    /* Azure doesn't support periods in function names */
     domain = await domainParser(req, domain, true);
 
     if (!domain) {
@@ -55,8 +54,7 @@ router.post('/:assistant_id', async (req, res) => {
     const action_id = _action_id ?? v4();
     const initialPromises = [];
 
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
 
     initialPromises.push(getAssistant({ assistant_id }));
     initialPromises.push(openai.beta.assistants.retrieve(assistant_id));
@@ -111,7 +109,7 @@ router.post('/:assistant_id', async (req, res) => {
     let updatedAssistant = await openai.beta.assistants.update(assistant_id, { tools });
     const promises = [];
     promises.push(
-      updateAssistant(
+      updateAssistantDoc(
         { assistant_id },
         {
           actions,
@@ -157,9 +155,7 @@ router.delete('/:assistant_id/:action_id/:model', async (req, res) => {
   try {
     const { assistant_id, action_id, model } = req.params;
     req.body.model = model;
-
-    /** @type {{ openai: OpenAI }} */
-    const { openai } = await initializeClient({ req, res });
+    const { openai } = await getOpenAIClient({ req, res });
 
     const initialPromises = [];
     initialPromises.push(getAssistant({ assistant_id }));
@@ -190,7 +186,7 @@ router.delete('/:assistant_id/:action_id/:model', async (req, res) => {
 
     const promises = [];
     promises.push(
-      updateAssistant(
+      updateAssistantDoc(
         { assistant_id },
         {
           actions: updatedActions,

api/server/routes/assistants/chatV1.js

@@ -0,0 +1,26 @@
+const express = require('express');
+
+const router = express.Router();
+const {
+  setHeaders,
+  handleAbort,
+  validateModel,
+  // validateEndpoint,
+  buildEndpointOption,
+} = require('~/server/middleware');
+const validateAssistant = require('~/server/middleware/assistants/validate');
+const chatController = require('~/server/controllers/assistants/chatV1');
+
+router.post('/abort', handleAbort());
+
+/**
+ * @route POST /
+ * @desc Chat with an assistant
+ * @access Public
+ * @param {express.Request} req - The request object, containing the request data.
+ * @param {express.Response} res - The response object, used to send back a response.
+ * @returns {void}
+ */
+router.post('/', validateModel, buildEndpointOption, validateAssistant, setHeaders, chatController);
+
+module.exports = router;

api/server/routes/assistants/chatV2.js

@@ -0,0 +1,26 @@
+const express = require('express');
+
+const router = express.Router();
+const {
+  setHeaders,
+  handleAbort,
+  validateModel,
+  // validateEndpoint,
+  buildEndpointOption,
+} = require('~/server/middleware');
+const validateAssistant = require('~/server/middleware/assistants/validate');
+const chatController = require('~/server/controllers/assistants/chatV2');
+
+router.post('/abort', handleAbort());
+
+/**
+ * @route POST /
+ * @desc Chat with an assistant
+ * @access Public
+ * @param {express.Request} req - The request object, containing the request data.
+ * @param {express.Response} res - The response object, used to send back a response.
+ * @returns {void}
+ */
+router.post('/', validateModel, buildEndpointOption, validateAssistant, setHeaders, chatController);
+
+module.exports = router;

api/server/routes/assistants/index.js

@@ -7,16 +7,19 @@ const {
   // concurrentLimiter,
   // messageIpLimiter,
   // messageUserLimiter,
-} = require('../../middleware');
+} = require('~/server/middleware');
 
-const assistants = require('./assistants');
-const chat = require('./chat');
+const v1 = require('./v1');
+const chatV1 = require('./chatV1');
+const v2 = require('./v2');
+const chatV2 = require('./chatV2');
 
 router.use(requireJwtAuth);
 router.use(checkBan);
 router.use(uaParser);
-
-router.use('/', assistants);
-router.use('/chat', chat);
+router.use('/v1/', v1);
+router.use('/v1/chat', chatV1);
+router.use('/v2/', v2);
+router.use('/v2/chat', chatV2);
 
 module.exports = router;

api/server/routes/assistants/v1.js

@@ -0,0 +1,81 @@
+const multer = require('multer');
+const express = require('express');
+const controllers = require('~/server/controllers/assistants/v1');
+const actions = require('./actions');
+const tools = require('./tools');
+
+const upload = multer();
+const router = express.Router();
+
+/**
+ * Assistant actions route.
+ * @route GET|POST /assistants/actions
+ */
+router.use('/actions', actions);
+
+/**
+ * Create an assistant.
+ * @route GET /assistants/tools
+ * @returns {TPlugin[]} 200 - application/json
+ */
+router.use('/tools', tools);
+
+/**
+ * Create an assistant.
+ * @route POST /assistants
+ * @param {AssistantCreateParams} req.body - The assistant creation parameters.
+ * @returns {Assistant} 201 - success response - application/json
+ */
+router.post('/', controllers.createAssistant);
+
+/**
+ * Retrieves an assistant.
+ * @route GET /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.get('/:id', controllers.retrieveAssistant);
+
+/**
+ * Modifies an assistant.
+ * @route PATCH /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @param {AssistantUpdateParams} req.body - The assistant update parameters.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.patch('/:id', controllers.patchAssistant);
+
+/**
+ * Deletes an assistant.
+ * @route DELETE /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.delete('/:id', controllers.deleteAssistant);
+
+/**
+ * Returns a list of assistants.
+ * @route GET /assistants
+ * @param {AssistantListParams} req.query - The assistant list parameters for pagination and sorting.
+ * @returns {AssistantListResponse} 200 - success response - application/json
+ */
+router.get('/', controllers.listAssistants);
+
+/**
+ * Returns a list of the user's assistant documents (metadata saved to database).
+ * @route GET /assistants/documents
+ * @returns {AssistantDocument[]} 200 - success response - application/json
+ */
+router.get('/documents', controllers.getAssistantDocuments);
+
+/**
+ * Uploads and updates an avatar for a specific assistant.
+ * @route POST /avatar/:assistant_id
+ * @param {string} req.params.assistant_id - The ID of the assistant.
+ * @param {Express.Multer.File} req.file - The avatar image file.
+ * @param {string} [req.body.metadata] - Optional metadata for the assistant's avatar.
+ * @returns {Object} 200 - success response - application/json
+ */
+router.post('/avatar/:assistant_id', upload.single('file'), controllers.uploadAssistantAvatar);
+
+module.exports = router;

api/server/routes/assistants/v2.js

@@ -0,0 +1,82 @@
+const multer = require('multer');
+const express = require('express');
+const v1 = require('~/server/controllers/assistants/v1');
+const v2 = require('~/server/controllers/assistants/v2');
+const actions = require('./actions');
+const tools = require('./tools');
+
+const upload = multer();
+const router = express.Router();
+
+/**
+ * Assistant actions route.
+ * @route GET|POST /assistants/actions
+ */
+router.use('/actions', actions);
+
+/**
+ * Create an assistant.
+ * @route GET /assistants/tools
+ * @returns {TPlugin[]} 200 - application/json
+ */
+router.use('/tools', tools);
+
+/**
+ * Create an assistant.
+ * @route POST /assistants
+ * @param {AssistantCreateParams} req.body - The assistant creation parameters.
+ * @returns {Assistant} 201 - success response - application/json
+ */
+router.post('/', v2.createAssistant);
+
+/**
+ * Retrieves an assistant.
+ * @route GET /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.get('/:id', v1.retrieveAssistant);
+
+/**
+ * Modifies an assistant.
+ * @route PATCH /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @param {AssistantUpdateParams} req.body - The assistant update parameters.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.patch('/:id', v2.patchAssistant);
+
+/**
+ * Deletes an assistant.
+ * @route DELETE /assistants/:id
+ * @param {string} req.params.id - Assistant identifier.
+ * @returns {Assistant} 200 - success response - application/json
+ */
+router.delete('/:id', v1.deleteAssistant);
+
+/**
+ * Returns a list of assistants.
+ * @route GET /assistants
+ * @param {AssistantListParams} req.query - The assistant list parameters for pagination and sorting.
+ * @returns {AssistantListResponse} 200 - success response - application/json
+ */
+router.get('/', v1.listAssistants);
+
+/**
+ * Returns a list of the user's assistant documents (metadata saved to database).
+ * @route GET /assistants/documents
+ * @returns {AssistantDocument[]} 200 - success response - application/json
+ */
+router.get('/documents', v1.getAssistantDocuments);
+
+/**
+ * Uploads and updates an avatar for a specific assistant.
+ * @route POST /avatar/:assistant_id
+ * @param {string} req.params.assistant_id - The ID of the assistant.
+ * @param {Express.Multer.File} req.file - The avatar image file.
+ * @param {string} [req.body.metadata] - Optional metadata for the assistant's avatar.
+ * @returns {Object} 200 - success response - application/json
+ */
+router.post('/avatar/:assistant_id', upload.single('file'), v1.uploadAssistantAvatar);
+
+module.exports = router;

api/server/routes/auth.js

@@ -1,29 +1,46 @@
 const express = require('express');
 const {
-  resetPasswordRequestController,
-  resetPasswordController,
   refreshController,
   registrationController,
-} = require('../controllers/AuthController');
-const { loginController } = require('../controllers/auth/LoginController');
-const { logoutController } = require('../controllers/auth/LogoutController');
+  resetPasswordController,
+  resetPasswordRequestController,
+} = require('~/server/controllers/AuthController');
+const { loginController } = require('~/server/controllers/auth/LoginController');
+const { logoutController } = require('~/server/controllers/auth/LogoutController');
 const {
   checkBan,
   loginLimiter,
-  registerLimiter,
   requireJwtAuth,
+  registerLimiter,
+  requireLdapAuth,
   requireLocalAuth,
+  resetPasswordLimiter,
   validateRegistration,
-} = require('../middleware');
+  validatePasswordReset,
+} = require('~/server/middleware');
 
 const router = express.Router();
 
+const ldapAuth =
+  !!process.env.LDAP_URL && !!process.env.LDAP_BIND_DN && !!process.env.LDAP_USER_SEARCH_BASE;
 //Local
 router.post('/logout', requireJwtAuth, logoutController);
-router.post('/login', loginLimiter, checkBan, requireLocalAuth, loginController);
+router.post(
+  '/login',
+  loginLimiter,
+  checkBan,
+  ldapAuth ? requireLdapAuth : requireLocalAuth,
+  loginController,
+);
 router.post('/refresh', refreshController);
 router.post('/register', registerLimiter, checkBan, validateRegistration, registrationController);
-router.post('/requestPasswordReset', resetPasswordRequestController);
-router.post('/resetPassword', resetPasswordController);
+router.post(
+  '/requestPasswordReset',
+  resetPasswordLimiter,
+  checkBan,
+  validatePasswordReset,
+  resetPasswordRequestController,
+);
+router.post('/resetPassword', checkBan, validatePasswordReset, resetPasswordController);
 
 module.exports = router;

api/server/routes/config.js

@@ -6,6 +6,15 @@ const { logger } = require('~/config');
 const router = express.Router();
 const emailLoginEnabled =
   process.env.ALLOW_EMAIL_LOGIN === undefined || isEnabled(process.env.ALLOW_EMAIL_LOGIN);
+const passwordResetEnabled = isEnabled(process.env.ALLOW_PASSWORD_RESET);
+
+const sharedLinksEnabled =
+  process.env.ALLOW_SHARED_LINKS === undefined || isEnabled(process.env.ALLOW_SHARED_LINKS);
+
+const publicSharedLinksEnabled =
+  sharedLinksEnabled &&
+  (process.env.ALLOW_SHARED_LINKS_PUBLIC === undefined ||
+    isEnabled(process.env.ALLOW_SHARED_LINKS_PUBLIC));
 
 router.get('/', async function (req, res) {
   const isBirthday = () => {
@@ -13,6 +22,8 @@ router.get('/', async function (req, res) {
     return today.getMonth() === 1 && today.getDate() === 11;
   };
 
+  const ldapLoginEnabled =
+    !!process.env.LDAP_URL && !!process.env.LDAP_BIND_DN && !!process.env.LDAP_USER_SEARCH_BASE;
   try {
     /** @type {TStartupConfig} */
     const payload = {
@@ -30,15 +41,17 @@ router.get('/', async function (req, res) {
         !!process.env.OPENID_SESSION_SECRET,
       openidLabel: process.env.OPENID_BUTTON_LABEL || 'Continue with OpenID',
       openidImageUrl: process.env.OPENID_IMAGE_URL,
+      ldapLoginEnabled,
       serverDomain: process.env.DOMAIN_SERVER || 'http://localhost:3080',
       emailLoginEnabled,
-      registrationEnabled: isEnabled(process.env.ALLOW_REGISTRATION),
+      registrationEnabled: !ldapLoginEnabled && isEnabled(process.env.ALLOW_REGISTRATION),
       socialLoginEnabled: isEnabled(process.env.ALLOW_SOCIAL_LOGIN),
       emailEnabled:
         (!!process.env.EMAIL_SERVICE || !!process.env.EMAIL_HOST) &&
         !!process.env.EMAIL_USERNAME &&
         !!process.env.EMAIL_PASSWORD &&
         !!process.env.EMAIL_FROM,
+      passwordResetEnabled,
       checkBalance: isEnabled(process.env.CHECK_BALANCE),
       showBirthdayIcon:
         isBirthday() ||
@@ -47,6 +60,9 @@ router.get('/', async function (req, res) {
       helpAndFaqURL: process.env.HELP_AND_FAQ_URL || 'https://librechat.ai',
       interface: req.app.locals.interfaceConfig,
       modelSpecs: req.app.locals.modelSpecs,
+      sharedLinksEnabled,
+      publicSharedLinksEnabled,
+      analyticsGtmId: process.env.ANALYTICS_GTM_ID,
     };
 
     if (typeof process.env.CUSTOM_FOOTER === 'string') {

api/server/routes/convos.js

@@ -3,12 +3,11 @@ const express = require('express');
 const { CacheKeys } = require('librechat-data-provider');
 const { initializeClient } = require('~/server/services/Endpoints/assistants');
 const { getConvosByPage, deleteConvos, getConvo, saveConvo } = require('~/models/Conversation');
-const { IMPORT_CONVERSATION_JOB_NAME } = require('~/server/utils/import/jobDefinition');
 const { storage, importFileFilter } = require('~/server/routes/files/multer');
 const requireJwtAuth = require('~/server/middleware/requireJwtAuth');
 const { forkConversation } = require('~/server/utils/import/fork');
+const { importConversations } = require('~/server/utils/import');
 const { createImportLimiters } = require('~/server/middleware');
-const jobScheduler = require('~/server/utils/jobScheduler');
 const getLogStores = require('~/cache/getLogStores');
 const { sleep } = require('~/server/utils');
 const { logger } = require('~/config');
@@ -129,10 +128,9 @@ router.post(
   upload.single('file'),
   async (req, res) => {
     try {
-      const filepath = req.file.path;
-      const job = await jobScheduler.now(IMPORT_CONVERSATION_JOB_NAME, filepath, req.user.id);
-
-      res.status(201).json({ message: 'Import started', jobId: job.id });
+      /* TODO: optimize to return imported conversations and add manually */
+      await importConversations({ filepath: req.file.path, requestUserId: req.user.id });
+      res.status(201).json({ message: 'Conversation(s) imported successfully' });
     } catch (error) {
       logger.error('Error processing file', error);
       res.status(500).send('Error processing file');
@@ -169,24 +167,4 @@ router.post('/fork', async (req, res) => {
   }
 });
 
-// Get the status of an import job for polling
-router.get('/import/jobs/:jobId', async (req, res) => {
-  try {
-    const { jobId } = req.params;
-    const { userId, ...jobStatus } = await jobScheduler.getJobStatus(jobId);
-    if (!jobStatus) {
-      return res.status(404).json({ message: 'Job not found.' });
-    }
-
-    if (userId !== req.user.id) {
-      return res.status(403).json({ message: 'Unauthorized' });
-    }
-
-    res.json(jobStatus);
-  } catch (error) {
-    logger.error('Error getting job details', error);
-    res.status(500).send('Error getting job details');
-  }
-});
-
 module.exports = router;

api/server/routes/edit/gptPlugins.js

@@ -110,7 +110,11 @@ router.post(
       if (!start) {
         saveMessage({ ...userMessage, user });
       }
-      sendIntermediateMessage(res, { plugin });
+      sendIntermediateMessage(res, {
+        plugin,
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
       // logger.debug('PLUGIN ACTION', formattedAction);
     };
 
@@ -119,7 +123,11 @@ router.post(
       plugin.outputs = steps && steps[0].action ? formatSteps(steps) : 'An error occurred.';
       plugin.loading = false;
       saveMessage({ ...userMessage, user });
-      sendIntermediateMessage(res, { plugin });
+      sendIntermediateMessage(res, {
+        plugin,
+        parentMessageId: userMessage.messageId,
+        messageId: responseMessageId,
+      });
       // logger.debug('CHAIN END', plugin.outputs);
     };
 
@@ -153,12 +161,13 @@ router.post(
         onChainEnd,
         onStart,
         ...endpointOption,
-        onProgress: progressCallback.call(null, {
+        progressCallback,
+        progressOptions: {
           res,
           text,
           plugin,
-          parentMessageId: overrideParentMessageId || userMessageId,
-        }),
+          // parentMessageId: overrideParentMessageId || userMessageId,
+        },
         abortController,
       });
 

api/server/routes/files/files.js

@@ -1,6 +1,6 @@
 const fs = require('fs').promises;
 const express = require('express');
-const { isUUID, FileSources } = require('librechat-data-provider');
+const { isUUID, checkOpenAIStorage } = require('librechat-data-provider');
 const {
   filterFile,
   processFileUpload,
@@ -89,7 +89,7 @@ router.get('/download/:userId/:file_id', async (req, res) => {
       return res.status(403).send('Forbidden');
     }
 
-    if (file.source === FileSources.openai && !file.model) {
+    if (checkOpenAIStorage(file.source) && !file.model) {
       logger.warn(`${errorPrefix} has no associated model: ${file_id}`);
       return res.status(400).send('The model used when creating this file is not available');
     }
@@ -110,7 +110,8 @@ router.get('/download/:userId/:file_id', async (req, res) => {
     let passThrough;
     /** @type {ReadableStream | undefined} */
     let fileStream;
-    if (file.source === FileSources.openai) {
+
+    if (checkOpenAIStorage(file.source)) {
       req.body = { model: file.model };
       const { openai } = await initializeClient({ req, res });
       logger.debug(`Downloading file ${file_id} from OpenAI`);

api/server/routes/files/index.js

@@ -1,10 +1,19 @@
 const express = require('express');
-const { uaParser, checkBan, requireJwtAuth, createFileLimiters } = require('~/server/middleware');
+const {
+  uaParser,
+  checkBan,
+  requireJwtAuth,
+  createFileLimiters,
+  createTTSLimiters,
+  createSTTLimiters,
+} = require('~/server/middleware');
 const { createMulterInstance } = require('./multer');
 
 const files = require('./files');
 const images = require('./images');
 const avatar = require('./avatar');
+const stt = require('./stt');
+const tts = require('./tts');
 
 const initialize = async () => {
   const router = express.Router();
@@ -12,6 +21,12 @@ const initialize = async () => {
   router.use(checkBan);
   router.use(uaParser);
 
+  /* Important: stt/tts routes must be added before the upload limiters */
+  const { sttIpLimiter, sttUserLimiter } = createSTTLimiters();
+  const { ttsIpLimiter, ttsUserLimiter } = createTTSLimiters();
+  router.use('/stt', sttIpLimiter, sttUserLimiter, stt);
+  router.use('/tts', ttsIpLimiter, ttsUserLimiter, tts);
+
   const upload = await createMulterInstance();
   const { fileUploadIpLimiter, fileUploadUserLimiter } = createFileLimiters();
   router.post('*', fileUploadIpLimiter, fileUploadUserLimiter);

api/server/routes/files/stt.js

@@ -0,0 +1,13 @@
+const express = require('express');
+const router = express.Router();
+const multer = require('multer');
+const { requireJwtAuth } = require('~/server/middleware/');
+const { speechToText } = require('~/server/services/Files/Audio');
+
+const upload = multer();
+
+router.post('/', requireJwtAuth, upload.single('audio'), async (req, res) => {
+  await speechToText(req, res);
+});
+
+module.exports = router;

api/server/routes/files/tts.js

@@ -0,0 +1,42 @@
+const multer = require('multer');
+const express = require('express');
+const { CacheKeys } = require('librechat-data-provider');
+const { getVoices, streamAudio, textToSpeech } = require('~/server/services/Files/Audio');
+const { getLogStores } = require('~/cache');
+const { logger } = require('~/config');
+
+const router = express.Router();
+const upload = multer();
+
+router.post('/manual', upload.none(), async (req, res) => {
+  await textToSpeech(req, res);
+});
+
+const logDebugMessage = (req, message) =>
+  logger.debug(`[streamAudio] user: ${req?.user?.id ?? 'UNDEFINED_USER'} | ${message}`);
+
+// TODO: test caching
+router.post('/', async (req, res) => {
+  try {
+    const audioRunsCache = getLogStores(CacheKeys.AUDIO_RUNS);
+    const audioRun = await audioRunsCache.get(req.body.runId);
+    logDebugMessage(req, 'start stream audio');
+    if (audioRun) {
+      logDebugMessage(req, 'stream audio already running');
+      return res.status(401).json({ error: 'Audio stream already running' });
+    }
+    audioRunsCache.set(req.body.runId, true);
+    await streamAudio(req, res);
+    logDebugMessage(req, 'end stream audio');
+    res.status(200).end();
+  } catch (error) {
+    logger.error(`[streamAudio] user: ${req.user.id} | Failed to stream audio: ${error}`);
+    res.status(500).json({ error: 'Failed to stream audio' });
+  }
+});
+
+router.get('/voices', async (req, res) => {
+  await getVoices(req, res);
+});
+
+module.exports = router;