Merge pull request #2376 from open-webui/dev

0.1.125

.env.example

@@ -10,7 +10,8 @@ OPENAI_API_KEY=''
 # DO NOT TRACK
 SCARF_NO_ANALYTICS=true
 DO_NOT_TRACK=true
+ANONYMIZED_TELEMETRY=false
 
 # Use locally bundled version of the LiteLLM cost map json
 # to avoid repetitive startup connections
-LITELLM_LOCAL_MODEL_COST_MAP="True"
+LITELLM_LOCAL_MODEL_COST_MAP="True"

.eslintrc.cjs

@@ -4,6 +4,7 @@ module.exports = {
 		'eslint:recommended',
 		'plugin:@typescript-eslint/recommended',
 		'plugin:svelte/recommended',
+		'plugin:cypress/recommended',
 		'prettier'
 	],
 	parser: '@typescript-eslint/parser',

.github/ISSUE_TEMPLATE/bug_report.md

@@ -24,6 +24,9 @@ assignees: ''
 
 ## Environment
 
+- **Open WebUI Version:** [e.g., 0.1.120]
+- **Ollama (if applicable):** [e.g., 0.1.30, 0.1.32-rc1]
+
 - **Operating System:** [e.g., Windows 10, macOS Big Sur, Ubuntu 20.04]
 - **Browser (if applicable):** [e.g., Chrome 100.0, Firefox 98.0]
 

.github/dependabot.yml

@@ -0,0 +1,11 @@
+version: 2
+updates:
+  - package-ecosystem: pip
+    directory: '/backend'
+    schedule:
+      interval: weekly
+  - package-ecosystem: 'github-actions'
+    directory: '/'
+    schedule:
+      # Check for updates to GitHub Actions every week
+      interval: 'weekly'

.github/pull_request_template.md

@@ -1,32 +1,72 @@
-## Pull Request Checklist
+# Pull Request Checklist
 
-- [ ] **Description:** Briefly describe the changes in this pull request.
+### Note to first-time contributors: Please open a discussion post in [Discussions](https://github.com/open-webui/open-webui/discussions) and describe your changes before submitting a pull request.
+
+**Before submitting, make sure you've checked the following:**
+
+- [ ] **Target branch:** Please verify that the pull request targets the `dev` branch.
+- [ ] **Description:** Provide a concise description of the changes made in this pull request.
 - [ ] **Changelog:** Ensure a changelog entry following the format of [Keep a Changelog](https://keepachangelog.com/) is added at the bottom of the PR description.
-- [ ] **Documentation:** Have you updated relevant documentation?
+- [ ] **Documentation:** Have you updated relevant documentation [Open WebUI Docs](https://github.com/open-webui/docs), or other documentation sources?
 - [ ] **Dependencies:** Are there any new dependencies? Have you updated the dependency versions in the documentation?
+- [ ] **Testing:** Have you written and run sufficient tests for validating the changes?
+- [ ] **Code review:** Have you performed a self-review of your code, addressing any coding standard issues and ensuring adherence to the project's coding standards?
+- [ ] **Label:** To cleary categorize this pull request, assign a relevant label to the pull request title, using one of the following:
+  - **BREAKING CHANGE**: Significant changes that may affect compatibility
+  - **build**: Changes that affect the build system or external dependencies
+  - **ci**: Changes to our continuous integration processes or workflows
+  - **chore**: Refactor, cleanup, or other non-functional code changes
+  - **docs**: Documentation update or addition
+  - **feat**: Introduces a new feature or enhancement to the codebase
+  - **fix**: Bug fix or error correction
+  - **i18n**: Internationalization or localization changes
+  - **perf**: Performance improvement
+  - **refactor**: Code restructuring for better maintainability, readability, or scalability
+  - **style**: Changes that do not affect the meaning of the code (white-space, formatting, missing semi-colons, etc.)
+  - **test**: Adding missing tests or correcting existing tests
+  - **WIP**: Work in progress, a temporary label for incomplete or ongoing work
 
----
+# Changelog Entry
 
-## Description
+### Description
 
-[Insert a brief description of the changes made in this pull request]
-
----
-
-### Changelog Entry
+- [Concisely describe the changes made in this pull request, including any relevant motivation and impact (e.g., fixing a bug, adding a feature, or improving performance)]
 
 ### Added
 
-- [List any new features or additions]
-
-### Fixed
-
-- [List any fixes or corrections]
+- [List any new features, functionalities, or additions]
 
 ### Changed
 
-- [List any changes or updates]
+- [List any changes, updates, refactorings, or optimizations]
+
+### Deprecated
+
+- [List any deprecated functionality or features that have been removed]
 
 ### Removed
 
-- [List any removed features or files]
+- [List any removed features, files, or functionalities]
+
+### Fixed
+
+- [List any fixes, corrections, or bug fixes]
+
+### Security
+
+- [List any new or updated security-related changes, including vulnerability fixes]
+
+### Breaking Changes
+
+- **BREAKING CHANGE**: [List any breaking changes affecting compatibility or functionality]
+
+---
+
+### Additional Information
+
+- [Insert any additional context, notes, or explanations for the changes]
+  - [Reference any related issues, commits, or other relevant information]
+
+### Screenshots or Videos
+
+- [Attach any relevant screenshots or videos demonstrating the changes]

.github/workflows/format-build-frontend.yaml

@@ -29,8 +29,29 @@ jobs:
       - name: Format Frontend
         run: npm run format
 
+      - name: Run i18next
+        run: npm run i18n:parse
+
       - name: Check for Changes After Format
         run: git diff --exit-code
 
       - name: Build Frontend
         run: npm run build
+
+  test-frontend:
+    name: 'Frontend Unit Tests'
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Install Dependencies
+        run: npm ci
+
+      - name: Run vitest
+        run: npm run test:frontend

.github/workflows/integration-test.yml

@@ -0,0 +1,195 @@
+name: Integration Test
+
+on:
+  push:
+    branches:
+      - main
+      - dev
+  pull_request:
+    branches:
+      - main
+      - dev
+
+jobs:
+  cypress-run:
+    name: Run Cypress Integration Tests
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v4
+
+      - name: Build and run Compose Stack
+        run: |
+          docker compose --file docker-compose.yaml --file docker-compose.api.yaml up --detach --build
+          
+      - name: Wait for Ollama to be up
+        timeout-minutes: 5
+        run: |
+          until curl --output /dev/null --silent --fail http://localhost:11434; do
+            printf '.'
+            sleep 1
+          done
+          echo "Service is up!"
+
+      - name: Preload Ollama model
+        run: |
+          docker exec ollama ollama pull qwen:0.5b-chat-v1.5-q2_K
+
+      - name: Cypress run
+        uses: cypress-io/github-action@v6
+        with:
+          browser: chrome
+          wait-on: 'http://localhost:3000'
+          config: baseUrl=http://localhost:3000
+
+      - uses: actions/upload-artifact@v4
+        if: always()
+        name: Upload Cypress videos
+        with:
+          name: cypress-videos
+          path: cypress/videos
+          if-no-files-found: ignore
+
+      - name: Extract Compose logs
+        if: always()
+        run: |
+          docker compose logs > compose-logs.txt
+
+      - uses: actions/upload-artifact@v4
+        if: always()
+        name: Upload Compose logs
+        with:
+          name: compose-logs
+          path: compose-logs.txt
+          if-no-files-found: ignore
+
+  migration_test:
+    name: Run Migration Tests
+    runs-on: ubuntu-latest
+    services:
+      postgres:
+        image: postgres
+        env:
+          POSTGRES_PASSWORD: postgres
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 5432:5432
+#      mysql:
+#        image: mysql
+#        env:
+#          MYSQL_ROOT_PASSWORD: mysql
+#          MYSQL_DATABASE: mysql
+#        options: >-
+#          --health-cmd "mysqladmin ping -h localhost"
+#          --health-interval 10s
+#          --health-timeout 5s
+#          --health-retries 5
+#        ports:
+#          - 3306:3306
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Set up uv
+        uses: yezz123/setup-uv@v4
+        with:
+          uv-venv: venv
+
+      - name: Activate virtualenv
+        run: |
+          . venv/bin/activate
+          echo PATH=$PATH >> $GITHUB_ENV
+
+      - name: Install dependencies
+        run: |
+          uv pip install -r backend/requirements.txt
+
+      - name: Test backend with SQLite
+        id: sqlite
+        env:
+          WEBUI_SECRET_KEY: secret-key
+          GLOBAL_LOG_LEVEL: debug
+        run: |
+          cd backend
+          uvicorn main:app --port "8080" --forwarded-allow-ips '*' &
+          UVICORN_PID=$!
+          # Wait up to 20 seconds for the server to start
+          for i in {1..20}; do
+              curl -s http://localhost:8080/api/config > /dev/null && break
+              sleep 1
+              if [ $i -eq 20 ]; then
+                  echo "Server failed to start"
+                  kill -9 $UVICORN_PID
+                  exit 1
+              fi
+          done
+          # Check that the server is still running after 5 seconds
+          sleep 5
+          if ! kill -0 $UVICORN_PID; then
+              echo "Server has stopped"
+              exit 1
+          fi
+          
+
+      - name: Test backend with Postgres
+        if: success() || steps.sqlite.conclusion == 'failure'
+        env:
+          WEBUI_SECRET_KEY: secret-key
+          GLOBAL_LOG_LEVEL: debug
+          DATABASE_URL: postgresql://postgres:postgres@localhost:5432/postgres
+        run: |
+          cd backend
+          uvicorn main:app --port "8081" --forwarded-allow-ips '*' &
+          UVICORN_PID=$!
+          # Wait up to 20 seconds for the server to start
+          for i in {1..20}; do
+              curl -s http://localhost:8081/api/config > /dev/null && break
+              sleep 1
+              if [ $i -eq 20 ]; then
+                  echo "Server failed to start"
+                  kill -9 $UVICORN_PID
+                  exit 1
+              fi
+          done
+          # Check that the server is still running after 5 seconds
+          sleep 5
+          if ! kill -0 $UVICORN_PID; then
+              echo "Server has stopped"
+              exit 1
+          fi
+
+#      - name: Test backend with MySQL
+#        if: success() || steps.sqlite.conclusion == 'failure' || steps.postgres.conclusion == 'failure'
+#        env:
+#          WEBUI_SECRET_KEY: secret-key
+#          GLOBAL_LOG_LEVEL: debug
+#          DATABASE_URL: mysql://root:mysql@localhost:3306/mysql
+#        run: |
+#          cd backend
+#          uvicorn main:app --port "8083" --forwarded-allow-ips '*' &
+#          UVICORN_PID=$!
+#          # Wait up to 20 seconds for the server to start
+#          for i in {1..20}; do
+#              curl -s http://localhost:8083/api/config > /dev/null && break
+#              sleep 1
+#              if [ $i -eq 20 ]; then
+#                  echo "Server failed to start"
+#                  kill -9 $UVICORN_PID
+#                  exit 1
+#              fi
+#          done
+#          # Check that the server is still running after 5 seconds
+#          sleep 5
+#          if ! kill -0 $UVICORN_PID; then
+#              echo "Server has stopped"
+#              exit 1
+#          fi

.gitignore

@@ -16,6 +16,10 @@ __pycache__/
 # C extensions
 *.so
 
+# Pyodide distribution
+static/pyodide/*
+!static/pyodide/pyodide-lock.json
+
 # Distribution / packaging
 .Python
 build/
@@ -297,4 +301,8 @@ dist
 .yarn/unplugged
 .yarn/build-state.yml
 .yarn/install-state.gz
-.pnp.*
+.pnp.*
+
+# cypress artifacts
+cypress/videos
+cypress/screenshots

.prettierignore

@@ -1,3 +1,11 @@
+# Ignore files for PNPM, NPM and YARN
+pnpm-lock.yaml
+package-lock.json
+yarn.lock
+
+kubernetes/
+
+# Copy of .gitignore
 .DS_Store
 node_modules
 /build
@@ -6,11 +14,303 @@ node_modules
 .env
 .env.*
 !.env.example
+vite.config.js.timestamp-*
+vite.config.ts.timestamp-*
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
 
-# Ignore files for PNPM, NPM and YARN
-pnpm-lock.yaml
-package-lock.json
-yarn.lock
+# C extensions
+*.so
 
-# Ignore kubernetes files
-kubernetes
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+.idea/
+
+# Logs
+logs
+*.log
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+lerna-debug.log*
+.pnpm-debug.log*
+
+# Diagnostic reports (https://nodejs.org/api/report.html)
+report.[0-9]*.[0-9]*.[0-9]*.[0-9]*.json
+
+# Runtime data
+pids
+*.pid
+*.seed
+*.pid.lock
+
+# Directory for instrumented libs generated by jscoverage/JSCover
+lib-cov
+
+# Coverage directory used by tools like istanbul
+coverage
+*.lcov
+
+# nyc test coverage
+.nyc_output
+
+# Grunt intermediate storage (https://gruntjs.com/creating-plugins#storing-task-files)
+.grunt
+
+# Bower dependency directory (https://bower.io/)
+bower_components
+
+# node-waf configuration
+.lock-wscript
+
+# Compiled binary addons (https://nodejs.org/api/addons.html)
+build/Release
+
+# Dependency directories
+node_modules/
+jspm_packages/
+
+# Snowpack dependency directory (https://snowpack.dev/)
+web_modules/
+
+# TypeScript cache
+*.tsbuildinfo
+
+# Optional npm cache directory
+.npm
+
+# Optional eslint cache
+.eslintcache
+
+# Optional stylelint cache
+.stylelintcache
+
+# Microbundle cache
+.rpt2_cache/
+.rts2_cache_cjs/
+.rts2_cache_es/
+.rts2_cache_umd/
+
+# Optional REPL history
+.node_repl_history
+
+# Output of 'npm pack'
+*.tgz
+
+# Yarn Integrity file
+.yarn-integrity
+
+# dotenv environment variable files
+.env
+.env.development.local
+.env.test.local
+.env.production.local
+.env.local
+
+# parcel-bundler cache (https://parceljs.org/)
+.cache
+.parcel-cache
+
+# Next.js build output
+.next
+out
+
+# Nuxt.js build / generate output
+.nuxt
+dist
+
+# Gatsby files
+.cache/
+# Comment in the public line in if your project uses Gatsby and not Next.js
+# https://nextjs.org/blog/next-9-1#public-directory-support
+# public
+
+# vuepress build output
+.vuepress/dist
+
+# vuepress v2.x temp and cache directory
+.temp
+.cache
+
+# Docusaurus cache and generated files
+.docusaurus
+
+# Serverless directories
+.serverless/
+
+# FuseBox cache
+.fusebox/
+
+# DynamoDB Local files
+.dynamodb/
+
+# TernJS port file
+.tern-port
+
+# Stores VSCode versions used for testing VSCode extensions
+.vscode-test
+
+# yarn v2
+.yarn/cache
+.yarn/unplugged
+.yarn/build-state.yml
+.yarn/install-state.gz
+.pnp.*
+
+# cypress artifacts
+cypress/videos
+cypress/screenshots
+
+
+
+/static/*

CHANGELOG.md

@@ -5,6 +5,146 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.125] - 2024-05-19
+
+### Added
+
+- **🔄 Updated UI**: Chat interface revamped with chat bubbles. Easily switch back to the old style via settings > interface > chat bubble UI.
+- **📂 Enhanced Sidebar UI**: Model files, documents, prompts, and playground merged into Workspace for streamlined access.
+- **🚀 Improved Many Model Interaction**: All responses now displayed simultaneously for a smoother experience.
+- **🐍 Python Code Execution**: Execute Python code locally in the browser with libraries like 'requests', 'beautifulsoup4', 'numpy', 'pandas', 'seaborn', 'matplotlib', 'scikit-learn', 'scipy', 'regex'.
+- **🧠 Experimental Memory Feature**: Manually input personal information you want LLMs to remember via settings > personalization > memory.
+- **💾 Persistent Settings**: Settings now saved as config.json for convenience.
+- **🩺 Health Check Endpoint**: Added for Docker deployment.
+- **↕️ RTL Support**: Toggle chat direction via settings > interface > chat direction.
+- **🖥️ PowerPoint Support**: RAG pipeline now supports PowerPoint documents.
+- **🌐 Language Updates**: Ukrainian, Turkish, Arabic, Chinese, Serbian, Vietnamese updated; Punjabi added.
+
+### Changed
+
+- **👤 Shared Chat Update**: Shared chat now includes creator user information.
+
+## [0.1.124] - 2024-05-08
+
+### Added
+
+- **🖼️ Improved Chat Sidebar**: Now conveniently displays time ranges and organizes chats by today, yesterday, and more.
+- **📜 Citations in RAG Feature**: Easily track the context fed to the LLM with added citations in the RAG feature.
+- **🔒 Auth Disable Option**: Introducing the ability to disable authentication. Set 'WEBUI_AUTH' to False to disable authentication. Note: Only applicable for fresh installations without existing users.
+- **📹 Enhanced YouTube RAG Pipeline**: Now supports non-English videos for an enriched experience.
+- **🔊 Specify OpenAI TTS Models**: Customize your TTS experience by specifying OpenAI TTS models.
+- **🔧 Additional Environment Variables**: Discover more environment variables in our comprehensive documentation at Open WebUI Documentation (https://docs.openwebui.com).
+- **🌐 Language Support**: Arabic, Finnish, and Hindi added; Improved support for German, Vietnamese, and Chinese.
+
+### Fixed
+
+- **🛠️ Model Selector Styling**: Addressed styling issues for improved user experience.
+- **⚠️ Warning Messages**: Resolved backend warning messages.
+
+### Changed
+
+- **📝 Title Generation**: Limited output to 50 tokens.
+- **📦 Helm Charts**: Removed Helm charts, now available in a separate repository (https://github.com/open-webui/helm-charts).
+
+## [0.1.123] - 2024-05-02
+
+### Added
+
+- **🎨 New Landing Page Design**: Refreshed design for a more modern look and optimized use of screen space.
+- **📹 Youtube RAG Pipeline**: Introduces dedicated RAG pipeline for Youtube videos, enabling interaction with video transcriptions directly.
+- **🔧 Enhanced Admin Panel**: Streamlined user management with options to add users directly or in bulk via CSV import.
+- **👥 '@' Model Integration**: Easily switch to specific models during conversations; old collaborative chat feature phased out.
+- **🌐 Language Enhancements**: Swedish translation added, plus improvements to German, Spanish, and the addition of Doge translation.
+
+### Fixed
+
+- **🗑️ Delete Chat Shortcut**: Addressed issue where shortcut wasn't functioning.
+- **🖼️ Modal Closing Bug**: Resolved unexpected closure of modal when dragging from within.
+- **✏️ Edit Button Styling**: Fixed styling inconsistency with edit buttons.
+- **🌐 Image Generation Compatibility Issue**: Rectified image generation compatibility issue with third-party APIs.
+- **📱 iOS PWA Icon Fix**: Corrected iOS PWA home screen icon shape.
+- **🔍 Scroll Gesture Bug**: Adjusted gesture sensitivity to prevent accidental activation when scrolling through code on mobile; now requires scrolling from the leftmost side to open the sidebar.
+
+### Changed
+
+- **🔄 Unlimited Context Length**: Advanced settings now allow unlimited max context length (previously limited to 16000).
+- **👑 Super Admin Assignment**: The first signup is automatically assigned a super admin role, unchangeable by other admins.
+- **🛡️ Admin User Restrictions**: User action buttons from the admin panel are now disabled for users with admin roles.
+- **🔝 Default Model Selector**: Set as default model option now exclusively available on the landing page.
+
+## [0.1.122] - 2024-04-27
+
+### Added
+
+- **🌟 Enhanced RAG Pipeline**: Now with hybrid searching via 'BM25', reranking powered by 'CrossEncoder', and configurable relevance score thresholds.
+- **🛢️ External Database Support**: Seamlessly connect to custom SQLite or Postgres databases using the 'DATABASE_URL' environment variable.
+- **🌐 Remote ChromaDB Support**: Introducing the capability to connect to remote ChromaDB servers.
+- **👨‍💼 Improved Admin Panel**: Admins can now conveniently check users' chat lists and last active status directly from the admin panel.
+- **🎨 Splash Screen**: Introducing a loading splash screen for a smoother user experience.
+- **🌍 Language Support Expansion**: Added support for Bangla (bn-BD), along with enhancements to Chinese, Spanish, and Ukrainian translations.
+- **💻 Improved LaTeX Rendering Performance**: Enjoy faster rendering times for LaTeX equations.
+- **🔧 More Environment Variables**: Explore additional environment variables in our documentation (https://docs.openwebui.com), including the 'ENABLE_LITELLM' option to manage memory usage.
+
+### Fixed
+
+- **🔧 Ollama Compatibility**: Resolved errors occurring when Ollama server version isn't an integer, such as SHA builds or RCs.
+- **🐛 Various OpenAI API Issues**: Addressed several issues related to the OpenAI API.
+- **🛑 Stop Sequence Issue**: Fixed the problem where the stop sequence with a backslash '\' was not functioning.
+- **🔤 Font Fallback**: Corrected font fallback issue.
+
+### Changed
+
+- **⌨️ Prompt Input Behavior on Mobile**: Enter key prompt submission disabled on mobile devices for improved user experience.
+
+## [0.1.121] - 2024-04-24
+
+### Fixed
+
+- **🔧 Translation Issues**: Addressed various translation discrepancies.
+- **🔒 LiteLLM Security Fix**: Updated LiteLLM version to resolve a security vulnerability.
+- **🖥️ HTML Tag Display**: Rectified the issue where the '< br >' tag wasn't displaying correctly.
+- **🔗 WebSocket Connection**: Resolved the failure of WebSocket connection under HTTPS security for ComfyUI server.
+- **📜 FileReader Optimization**: Implemented FileReader initialization per image in multi-file drag & drop to ensure reusability.
+- **🏷️ Tag Display**: Corrected tag display inconsistencies.
+- **📦 Archived Chat Styling**: Fixed styling issues in archived chat.
+- **🔖 Safari Copy Button Bug**: Addressed the bug where the copy button failed to copy links in Safari.
+
+## [0.1.120] - 2024-04-20
+
+### Added
+
+- **📦 Archive Chat Feature**: Easily archive chats with a new sidebar button, and access archived chats via the profile button > archived chats.
+- **🔊 Configurable Text-to-Speech Endpoint**: Customize your Text-to-Speech experience with configurable OpenAI endpoints.
+- **🛠️ Improved Error Handling**: Enhanced error message handling for connection failures.
+- **⌨️ Enhanced Shortcut**: When editing messages, use ctrl/cmd+enter to save and submit, and esc to close.
+- **🌐 Language Support**: Added support for Georgian and enhanced translations for Portuguese and Vietnamese.
+
+### Fixed
+
+- **🔧 Model Selector**: Resolved issue where default model selection was not saving.
+- **🔗 Share Link Copy Button**: Fixed bug where the copy button wasn't copying links in Safari.
+- **🎨 Light Theme Styling**: Addressed styling issue with the light theme.
+
+## [0.1.119] - 2024-04-16
+
+### Added
+
+- **🌟 Enhanced RAG Embedding Support**: Ollama, and OpenAI models can now be used for RAG embedding model.
+- **🔄 Seamless Integration**: Copy 'ollama run <model name>' directly from Ollama page to easily select and pull models.
+- **🏷️ Tagging Feature**: Add tags to chats directly via the sidebar chat menu.
+- **📱 Mobile Accessibility**: Swipe left and right on mobile to effortlessly open and close the sidebar.
+- **🔍 Improved Navigation**: Admin panel now supports pagination for user list.
+- **🌍 Additional Language Support**: Added Polish language support.
+
+### Fixed
+
+- **🌍 Language Enhancements**: Vietnamese and Spanish translations have been improved.
+- **🔧 Helm Fixes**: Resolved issues with Helm trailing slash and manifest.json.
+
+### Changed
+
+- **🐳 Docker Optimization**: Updated docker image build process to utilize 'uv' for significantly faster builds compared to 'pip3'.
+
 ## [0.1.118] - 2024-04-10
 
 ### Added

Dockerfile

@@ -8,8 +8,12 @@ ARG USE_CUDA_VER=cu121
 # any sentence transformer model; models to use can be found at https://huggingface.co/models?library=sentence-transformers
 # Leaderboard: https://huggingface.co/spaces/mteb/leaderboard 
 # for better performance and multilangauge support use "intfloat/multilingual-e5-large" (~2.5GB) or "intfloat/multilingual-e5-base" (~1.5GB)
-# IMPORTANT: If you change the default model (all-MiniLM-L6-v2) and vice versa, you aren't able to use RAG Chat with your previous documents loaded in the WebUI! You need to re-embed them.
-ARG USE_EMBEDDING_MODEL=all-MiniLM-L6-v2
+# IMPORTANT: If you change the embedding model (sentence-transformers/all-MiniLM-L6-v2) and vice versa, you aren't able to use RAG Chat with your previous documents loaded in the WebUI! You need to re-embed them.
+ARG USE_EMBEDDING_MODEL=sentence-transformers/all-MiniLM-L6-v2
+ARG USE_RERANKING_MODEL=""
+# Override at your own risk - non-root configurations are untested
+ARG UID=0
+ARG GID=0
 
 ######## WebUI frontend ########
 FROM --platform=$BUILDPLATFORM node:21-alpine3.19 as build
@@ -30,6 +34,9 @@ ARG USE_CUDA
 ARG USE_OLLAMA
 ARG USE_CUDA_VER
 ARG USE_EMBEDDING_MODEL
+ARG USE_RERANKING_MODEL
+ARG UID
+ARG GID
 
 ## Basis ##
 ENV ENV=prod \
@@ -38,7 +45,8 @@ ENV ENV=prod \
     USE_OLLAMA_DOCKER=${USE_OLLAMA} \
     USE_CUDA_DOCKER=${USE_CUDA} \
     USE_CUDA_DOCKER_VER=${USE_CUDA_VER} \
-    USE_EMBEDDING_MODEL_DOCKER=${USE_EMBEDDING_MODEL}
+    USE_EMBEDDING_MODEL_DOCKER=${USE_EMBEDDING_MODEL} \
+    USE_RERANKING_MODEL_DOCKER=${USE_RERANKING_MODEL}
 
 ## Basis URL Config ##
 ENV OLLAMA_BASE_URL="/ollama" \
@@ -48,7 +56,8 @@ ENV OLLAMA_BASE_URL="/ollama" \
 ENV OPENAI_API_KEY="" \
     WEBUI_SECRET_KEY="" \
     SCARF_NO_ANALYTICS=true \
-    DO_NOT_TRACK=true
+    DO_NOT_TRACK=true \
+    ANONYMIZED_TELEMETRY=false
 
 # Use locally bundled version of the LiteLLM cost map json
 # to avoid repetitive startup connections
@@ -62,48 +71,67 @@ ENV WHISPER_MODEL="base" \
 
 ## RAG Embedding model settings ##
 ENV RAG_EMBEDDING_MODEL="$USE_EMBEDDING_MODEL_DOCKER" \
-    RAG_EMBEDDING_MODEL_DIR="/app/backend/data/cache/embedding/models" \
+    RAG_RERANKING_MODEL="$USE_RERANKING_MODEL_DOCKER" \
     SENTENCE_TRANSFORMERS_HOME="/app/backend/data/cache/embedding/models"
+
+## Hugging Face download cache ##
+ENV HF_HOME="/app/backend/data/cache/embedding/models"
 #### Other models ##########################################################
 
 WORKDIR /app/backend
 
+ENV HOME /root
+# Create user and group if not root
+RUN if [ $UID -ne 0 ]; then \
+      if [ $GID -ne 0 ]; then \
+        addgroup --gid $GID app; \
+      fi; \
+      adduser --uid $UID --gid $GID --home $HOME --disabled-password --no-create-home app; \
+    fi
+
+RUN mkdir -p $HOME/.cache/chroma
+RUN echo -n 00000000-0000-0000-0000-000000000000 > $HOME/.cache/chroma/telemetry_user_id
+
+# Make sure the user has access to the app and root directory
+RUN chown -R $UID:$GID /app $HOME
+
 RUN if [ "$USE_OLLAMA" = "true" ]; then \
-        apt-get update && \
-        # Install pandoc and netcat
-        apt-get install -y --no-install-recommends pandoc netcat-openbsd && \
-        # for RAG OCR
-        apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
-        # install helper tools
-        apt-get install -y --no-install-recommends curl && \
-        # install ollama
-        curl -fsSL https://ollama.com/install.sh | sh && \
-        # cleanup
-        rm -rf /var/lib/apt/lists/*; \
+    apt-get update && \
+    # Install pandoc and netcat
+    apt-get install -y --no-install-recommends pandoc netcat-openbsd curl && \
+    # for RAG OCR
+    apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
+    # install helper tools
+    apt-get install -y --no-install-recommends curl jq && \
+    # install ollama
+    curl -fsSL https://ollama.com/install.sh | sh && \
+    # cleanup
+    rm -rf /var/lib/apt/lists/*; \
     else \
-        apt-get update && \
-        # Install pandoc and netcat
-        apt-get install -y --no-install-recommends pandoc netcat-openbsd && \
-        # for RAG OCR
-        apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
-        # cleanup
-        rm -rf /var/lib/apt/lists/*; \
+    apt-get update && \
+    # Install pandoc and netcat
+    apt-get install -y --no-install-recommends pandoc netcat-openbsd curl jq && \
+    # for RAG OCR
+    apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
+    # cleanup
+    rm -rf /var/lib/apt/lists/*; \
     fi
 
 # install python dependencies
-COPY ./backend/requirements.txt ./requirements.txt
+COPY --chown=$UID:$GID ./backend/requirements.txt ./requirements.txt
 
-RUN if [ "$USE_CUDA" = "true" ]; then \
-        # If you use CUDA the whisper and embedding model will be downloaded on first use
-        pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/$USE_CUDA_DOCKER_VER --no-cache-dir && \
-        pip3 install -r requirements.txt --no-cache-dir && \
-        python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='cpu', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])" && \
-        python -c "import os; from chromadb.utils import embedding_functions; sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name=os.environ['RAG_EMBEDDING_MODEL'], device='cpu')"; \
+RUN pip3 install uv && \
+    if [ "$USE_CUDA" = "true" ]; then \
+    # If you use CUDA the whisper and embedding model will be downloaded on first use
+    pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/$USE_CUDA_DOCKER_VER --no-cache-dir && \
+    uv pip install --system -r requirements.txt --no-cache-dir && \
+    python -c "import os; from sentence_transformers import SentenceTransformer; SentenceTransformer(os.environ['RAG_EMBEDDING_MODEL'], device='cpu')" && \
+    python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='cpu', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])"; \
     else \
-        pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu --no-cache-dir && \
-        pip3 install -r requirements.txt --no-cache-dir && \
-        python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='cpu', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])" && \
-        python -c "import os; from chromadb.utils import embedding_functions; sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name=os.environ['RAG_EMBEDDING_MODEL'], device='cpu')"; \
+    pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu --no-cache-dir && \
+    uv pip install --system -r requirements.txt --no-cache-dir && \
+    python -c "import os; from sentence_transformers import SentenceTransformer; SentenceTransformer(os.environ['RAG_EMBEDDING_MODEL'], device='cpu')" && \
+    python -c "import os; from faster_whisper import WhisperModel; WhisperModel(os.environ['WHISPER_MODEL'], device='cpu', compute_type='int8', download_root=os.environ['WHISPER_MODEL_DIR'])"; \
     fi
 
 
@@ -113,13 +141,17 @@ RUN if [ "$USE_CUDA" = "true" ]; then \
 # COPY --from=build /app/onnx /root/.cache/chroma/onnx_models/all-MiniLM-L6-v2/onnx
 
 # copy built frontend files
-COPY --from=build /app/build /app/build
-COPY --from=build /app/CHANGELOG.md /app/CHANGELOG.md
-COPY --from=build /app/package.json /app/package.json
+COPY --chown=$UID:$GID --from=build /app/build /app/build
+COPY --chown=$UID:$GID --from=build /app/CHANGELOG.md /app/CHANGELOG.md
+COPY --chown=$UID:$GID --from=build /app/package.json /app/package.json
 
 # copy backend files
-COPY ./backend .
+COPY --chown=$UID:$GID ./backend .
 
 EXPOSE 8080
 
-CMD [ "bash", "start.sh"]
+HEALTHCHECK CMD curl --silent --fail http://localhost:8080/health | jq -e '.status == true' || exit 1
+
+USER $UID:$GID
+
+CMD [ "bash", "start.sh"]

Makefile

@@ -1,27 +1,33 @@
+
+ifneq ($(shell which docker-compose 2>/dev/null),)
+    DOCKER_COMPOSE := docker-compose
+else
+    DOCKER_COMPOSE := docker compose
+endif
+
 install:
-	@docker-compose up -d
+	$(DOCKER_COMPOSE) up -d
 
 remove:
 	@chmod +x confirm_remove.sh
 	@./confirm_remove.sh
 
-
 start:
-	@docker-compose start
+	$(DOCKER_COMPOSE) start
 startAndBuild: 
-	docker-compose up -d --build
+	$(DOCKER_COMPOSE) up -d --build
 
 stop:
-	@docker-compose stop
+	$(DOCKER_COMPOSE) stop
 
 update:
 	# Calls the LLM update script
 	chmod +x update_ollama_models.sh
 	@./update_ollama_models.sh
 	@git pull
-	@docker-compose down
+	$(DOCKER_COMPOSE) down
 	# Make sure the ollama-webui container is stopped before rebuilding
 	@docker stop open-webui || true
-	@docker-compose up --build -d
-	@docker-compose start
+	$(DOCKER_COMPOSE) up --build -d
+	$(DOCKER_COMPOSE) start
 

README.md

@@ -25,22 +25,28 @@ Open WebUI is an extensible, feature-rich, and user-friendly self-hosted WebUI d
 
 - 🚀 **Effortless Setup**: Install seamlessly using Docker or Kubernetes (kubectl, kustomize or helm) for a hassle-free experience.
 
+- 🌈 **Theme Customization**: Choose from a variety of themes to personalize your Open WebUI experience.
+
 - 💻 **Code Syntax Highlighting**: Enjoy enhanced code readability with our syntax highlighting feature.
 
 - ✒️🔢 **Full Markdown and LaTeX Support**: Elevate your LLM experience with comprehensive Markdown and LaTeX capabilities for enriched interaction.
 
 - 📚 **Local RAG Integration**: Dive into the future of chat interactions with the groundbreaking Retrieval Augmented Generation (RAG) support. This feature seamlessly integrates document interactions into your chat experience. You can load documents directly into the chat or add files to your document library, effortlessly accessing them using `#` command in the prompt. In its alpha phase, occasional issues may arise as we actively refine and enhance this feature to ensure optimal performance and reliability.
 
+- 🔍 **RAG Embedding Support**: Change the RAG embedding model directly in document settings, enhancing document processing. This feature supports Ollama and OpenAI models.
+
 - 🌐 **Web Browsing Capability**: Seamlessly integrate websites into your chat experience using the `#` command followed by the URL. This feature allows you to incorporate web content directly into your conversations, enhancing the richness and depth of your interactions.
 
 - 📜 **Prompt Preset Support**: Instantly access preset prompts using the `/` command in the chat input. Load predefined conversation starters effortlessly and expedite your interactions. Effortlessly import prompts through [Open WebUI Community](https://openwebui.com/) integration.
 
-- 👍👎 **RLHF Annotation**: Empower your messages by rating them with thumbs up and thumbs down, facilitating the creation of datasets for Reinforcement Learning from Human Feedback (RLHF). Utilize your messages to train or fine-tune models, all while ensuring the confidentiality of locally saved data.
+- 👍👎 **RLHF Annotation**: Empower your messages by rating them with thumbs up and thumbs down, followed by the option to provide textual feedback, facilitating the creation of datasets for Reinforcement Learning from Human Feedback (RLHF). Utilize your messages to train or fine-tune models, all while ensuring the confidentiality of locally saved data.
 
 - 🏷️ **Conversation Tagging**: Effortlessly categorize and locate specific chats for quick reference and streamlined data collection.
 
 - 📥🗑️ **Download/Delete Models**: Easily download or remove models directly from the web UI.
 
+- 🔄 **Update All Ollama Models**: Easily update locally installed models all at once with a convenient button, streamlining model management.
+
 - ⬆️ **GGUF File Model Creation**: Effortlessly create Ollama models by uploading GGUF files directly from the web UI. Streamlined process with options to upload from your machine or download GGUF files from Hugging Face.
 
 - 🤖 **Multiple Model Support**: Seamlessly switch between different chat models for diverse interactions.
@@ -53,28 +59,42 @@ Open WebUI is an extensible, feature-rich, and user-friendly self-hosted WebUI d
 
 - 💬 **Collaborative Chat**: Harness the collective intelligence of multiple models by seamlessly orchestrating group conversations. Use the `@` command to specify the model, enabling dynamic and diverse dialogues within your chat interface. Immerse yourself in the collective intelligence woven into your chat environment.
 
+- 🗨️ **Local Chat Sharing**: Generate and share chat links seamlessly between users, enhancing collaboration and communication.
+
 - 🔄 **Regeneration History Access**: Easily revisit and explore your entire regeneration history.
 
 - 📜 **Chat History**: Effortlessly access and manage your conversation history.
 
+- 📬 **Archive Chats**: Effortlessly store away completed conversations with LLMs for future reference, maintaining a tidy and clutter-free chat interface while allowing for easy retrieval and reference.
+
 - 📤📥 **Import/Export Chat History**: Seamlessly move your chat data in and out of the platform.
 
 - 🗣️ **Voice Input Support**: Engage with your model through voice interactions; enjoy the convenience of talking to your model directly. Additionally, explore the option for sending voice input automatically after 3 seconds of silence for a streamlined experience.
 
+- 🔊 **Configurable Text-to-Speech Endpoint**: Customize your Text-to-Speech experience with configurable OpenAI endpoints.
+
 - ⚙️ **Fine-Tuned Control with Advanced Parameters**: Gain a deeper level of control by adjusting parameters such as temperature and defining your system prompts to tailor the conversation to your specific preferences and needs.
 
-- 🎨🤖 **Image Generation Integration**: Seamlessly incorporate image generation capabilities using AUTOMATIC1111 API (local) and DALL-E, enriching your chat experience with dynamic visual content.
+- 🎨🤖 **Image Generation Integration**: Seamlessly incorporate image generation capabilities using options such as AUTOMATIC1111 API (local), ComfyUI (local), and DALL-E, enriching your chat experience with dynamic visual content.
 
 - 🤝 **OpenAI API Integration**: Effortlessly integrate OpenAI-compatible API for versatile conversations alongside Ollama models. Customize the API Base URL to link with **LMStudio, Mistral, OpenRouter, and more**.
 
 - ✨ **Multiple OpenAI-Compatible API Support**: Seamlessly integrate and customize various OpenAI-compatible APIs, enhancing the versatility of your chat interactions.
 
+- 🔑 **API Key Generation Support**: Generate secret keys to leverage Open WebUI with OpenAI libraries, simplifying integration and development.
+
 - 🔗 **External Ollama Server Connection**: Seamlessly link to an external Ollama server hosted on a different address by configuring the environment variable.
 
 - 🔀 **Multiple Ollama Instance Load Balancing**: Effortlessly distribute chat requests across multiple Ollama instances for enhanced performance and reliability.
 
 - 👥 **Multi-User Management**: Easily oversee and administer users via our intuitive admin panel, streamlining user management processes.
 
+- 🔗 **Webhook Integration**: Subscribe to new user sign-up events via webhook (compatible with Google Chat and Microsoft Teams), providing real-time notifications and automation capabilities.
+
+- 🛡️ **Model Whitelisting**: Admins can whitelist models for users with the 'user' role, enhancing security and access control.
+
+- 📧 **Trusted Email Authentication**: Authenticate using a trusted email header, adding an additional layer of security and authentication.
+
 - 🔐 **Role-Based Access Control (RBAC)**: Ensure secure access with restricted permissions; only authorized individuals can access your Ollama, and exclusive model creation/pulling rights are reserved for administrators.
 
 - 🔒 **Backend Reverse Proxy Support**: Bolster security through direct communication between Open WebUI backend and Ollama. This key feature eliminates the need to expose Ollama over LAN. Requests made to the '/ollama/api' route from the web UI are seamlessly redirected to Ollama from the backend, enhancing overall system security.
@@ -100,22 +120,66 @@ Don't forget to explore our sibling project, [Open WebUI Community](https://open
 > [!TIP]  
 > If you wish to utilize Open WebUI with Ollama included or CUDA acceleration, we recommend utilizing our official images tagged with either `:cuda` or `:ollama`. To enable CUDA, you must install the [Nvidia CUDA container toolkit](https://docs.nvidia.com/dgx/nvidia-container-runtime-upgrade/) on your Linux/WSL system.
 
-**If Ollama is on your computer**, use this command:
+### Installation with Default Configuration
 
-```bash
-docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
-```
+- **If Ollama is on your computer**, use this command:
 
-**If Ollama is on a Different Server**, use this command:
+  ```bash
+  docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
+  ```
 
-To connect to Ollama on another server, change the `OLLAMA_BASE_URL` to the server's URL:
+- **If Ollama is on a Different Server**, use this command:
 
-```bash
-docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=https://example.com -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
-```
+  To connect to Ollama on another server, change the `OLLAMA_BASE_URL` to the server's URL:
+
+  ```bash
+  docker run -d -p 3000:8080 -e OLLAMA_BASE_URL=https://example.com -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
+  ```
+
+  - **To run Open WebUI with Nvidia GPU support**, use this command:
+
+  ```bash
+  docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda
+  ```
+
+### Installation for OpenAI API Usage Only
+
+- **If you're only using OpenAI API**, use this command:
+
+  ```bash
+  docker run -d -p 3000:8080 -e OPENAI_API_KEY=your_secret_key -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
+  ```
+
+### Installing Open WebUI with Bundled Ollama Support
+
+This installation method uses a single container image that bundles Open WebUI with Ollama, allowing for a streamlined setup via a single command. Choose the appropriate command based on your hardware setup:
+
+- **With GPU Support**:
+  Utilize GPU resources by running the following command:
+
+  ```bash
+  docker run -d -p 3000:8080 --gpus=all -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama
+  ```
+
+- **For CPU Only**:
+  If you're not using a GPU, use this command instead:
+
+  ```bash
+  docker run -d -p 3000:8080 -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama
+  ```
+
+Both commands facilitate a built-in, hassle-free installation of both Open WebUI and Ollama, ensuring that you can get everything up and running swiftly.
 
 After installation, you can access Open WebUI at [http://localhost:3000](http://localhost:3000). Enjoy! 😄
 
+### Other Installation Methods
+
+We offer various installation alternatives, including non-Docker native installation methods, Docker Compose, Kustomize, and Helm. Visit our [Open WebUI Documentation](https://docs.openwebui.com/getting-started/) or join our [Discord community](https://discord.gg/5rJgQTnV4s) for comprehensive guidance.
+
+### Troubleshooting
+
+Encountering connection issues? Our [Open WebUI Documentation](https://docs.openwebui.com/troubleshooting/) has got you covered. For further assistance and to join our vibrant community, visit the [Open WebUI Discord](https://discord.gg/5rJgQTnV4s).
+
 #### Open WebUI: Server Connection Error
 
 If you're experiencing connection issues, it’s often due to the WebUI docker container not being able to reach the Ollama server at 127.0.0.1:11434 (host.docker.internal:11434) inside the container . Use the `--network=host` flag in your docker command to resolve this. Note that the port changes from 3000 to 8080, resulting in the link: `http://localhost:8080`.
@@ -126,14 +190,6 @@ If you're experiencing connection issues, it’s often due to the WebUI docker c
 docker run -d --network=host -v open-webui:/app/backend/data -e OLLAMA_BASE_URL=http://127.0.0.1:11434 --name open-webui --restart always ghcr.io/open-webui/open-webui:main
 ```
 
-### Other Installation Methods
-
-We offer various installation alternatives, including non-Docker methods, Docker Compose, Kustomize, and Helm. Visit our [Open WebUI Documentation](https://docs.openwebui.com/getting-started/) or join our [Discord community](https://discord.gg/5rJgQTnV4s) for comprehensive guidance.
-
-### Troubleshooting
-
-Encountering connection issues? Our [Open WebUI Documentation](https://docs.openwebui.com/getting-started/troubleshooting/) has got you covered. For further assistance and to join our vibrant community, visit the [Open WebUI Discord](https://discord.gg/5rJgQTnV4s).
-
 ### Keeping Your Docker Installation Up-to-Date
 
 In case you want to update your local Docker installation to the latest version, you can do it with [Watchtower](https://containrrr.dev/watchtower/):
@@ -185,4 +241,4 @@ If you have any questions, suggestions, or need assistance, please open an issue
 
 ---
 
-Created by [Timothy J. Baek](https://github.com/tjbck) - Let's make Open Web UI even more amazing together! 💪
+Created by [Timothy J. Baek](https://github.com/tjbck) - Let's make Open WebUI even more amazing together! 💪

backend/apps/audio/main.py

@@ -10,8 +10,19 @@ from fastapi import (
     File,
     Form,
 )
+
+from fastapi.responses import StreamingResponse, JSONResponse, FileResponse
+
 from fastapi.middleware.cors import CORSMiddleware
 from faster_whisper import WhisperModel
+from pydantic import BaseModel
+
+
+import requests
+import hashlib
+from pathlib import Path
+import json
+
 
 from constants import ERROR_MESSAGES
 from utils.utils import (
@@ -28,7 +39,13 @@ from config import (
     UPLOAD_DIR,
     WHISPER_MODEL,
     WHISPER_MODEL_DIR,
+    WHISPER_MODEL_AUTO_UPDATE,
     DEVICE_TYPE,
+    AUDIO_OPENAI_API_BASE_URL,
+    AUDIO_OPENAI_API_KEY,
+    AUDIO_OPENAI_API_MODEL,
+    AUDIO_OPENAI_API_VOICE,
+    AppConfig,
 )
 
 log = logging.getLogger(__name__)
@@ -43,12 +60,114 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+app.state.config = AppConfig()
+app.state.config.OPENAI_API_BASE_URL = AUDIO_OPENAI_API_BASE_URL
+app.state.config.OPENAI_API_KEY = AUDIO_OPENAI_API_KEY
+app.state.config.OPENAI_API_MODEL = AUDIO_OPENAI_API_MODEL
+app.state.config.OPENAI_API_VOICE = AUDIO_OPENAI_API_VOICE
+
 # setting device type for whisper model
 whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu"
 log.info(f"whisper_device_type: {whisper_device_type}")
 
+SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
+SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
 
-@app.post("/transcribe")
+
+class OpenAIConfigUpdateForm(BaseModel):
+    url: str
+    key: str
+    model: str
+    speaker: str
+
+
+@app.get("/config")
+async def get_openai_config(user=Depends(get_admin_user)):
+    return {
+        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+        "OPENAI_API_MODEL": app.state.config.OPENAI_API_MODEL,
+        "OPENAI_API_VOICE": app.state.config.OPENAI_API_VOICE,
+    }
+
+
+@app.post("/config/update")
+async def update_openai_config(
+    form_data: OpenAIConfigUpdateForm, user=Depends(get_admin_user)
+):
+    if form_data.key == "":
+        raise HTTPException(status_code=400, detail=ERROR_MESSAGES.API_KEY_NOT_FOUND)
+
+    app.state.config.OPENAI_API_BASE_URL = form_data.url
+    app.state.config.OPENAI_API_KEY = form_data.key
+    app.state.config.OPENAI_API_MODEL = form_data.model
+    app.state.config.OPENAI_API_VOICE = form_data.speaker
+
+    return {
+        "status": True,
+        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+        "OPENAI_API_MODEL": app.state.config.OPENAI_API_MODEL,
+        "OPENAI_API_VOICE": app.state.config.OPENAI_API_VOICE,
+    }
+
+
+@app.post("/speech")
+async def speech(request: Request, user=Depends(get_verified_user)):
+    body = await request.body()
+    name = hashlib.sha256(body).hexdigest()
+
+    file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
+    file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
+
+    # Check if the file already exists in the cache
+    if file_path.is_file():
+        return FileResponse(file_path)
+
+    headers = {}
+    headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}"
+    headers["Content-Type"] = "application/json"
+
+    r = None
+    try:
+        r = requests.post(
+            url=f"{app.state.config.OPENAI_API_BASE_URL}/audio/speech",
+            data=body,
+            headers=headers,
+            stream=True,
+        )
+
+        r.raise_for_status()
+
+        # Save the streaming content to a file
+        with open(file_path, "wb") as f:
+            for chunk in r.iter_content(chunk_size=8192):
+                f.write(chunk)
+
+        with open(file_body_path, "w") as f:
+            json.dump(json.loads(body.decode("utf-8")), f)
+
+        # Return the saved file
+        return FileResponse(file_path)
+
+    except Exception as e:
+        log.exception(e)
+        error_detail = "Open WebUI: Server Connection Error"
+        if r is not None:
+            try:
+                res = r.json()
+                if "error" in res:
+                    error_detail = f"External: {res['error']['message']}"
+            except:
+                error_detail = f"External: {e}"
+
+        raise HTTPException(
+            status_code=r.status_code if r != None else 500,
+            detail=error_detail,
+        )
+
+
+@app.post("/transcriptions")
 def transcribe(
     file: UploadFile = File(...),
     user=Depends(get_current_user),
@@ -69,12 +188,24 @@ def transcribe(
             f.write(contents)
             f.close()
 
-        model = WhisperModel(
-            WHISPER_MODEL,
-            device=whisper_device_type,
-            compute_type="int8",
-            download_root=WHISPER_MODEL_DIR,
-        )
+        whisper_kwargs = {
+            "model_size_or_path": WHISPER_MODEL,
+            "device": whisper_device_type,
+            "compute_type": "int8",
+            "download_root": WHISPER_MODEL_DIR,
+            "local_files_only": not WHISPER_MODEL_AUTO_UPDATE,
+        }
+
+        log.debug(f"whisper_kwargs: {whisper_kwargs}")
+
+        try:
+            model = WhisperModel(**whisper_kwargs)
+        except:
+            log.warning(
+                "WhisperModel initialization failed, attempting download with local_files_only=False"
+            )
+            whisper_kwargs["local_files_only"] = False
+            model = WhisperModel(**whisper_kwargs)
 
         segments, info = model.transcribe(file_path, beam_size=5)
         log.info(

backend/apps/images/main.py

@@ -24,12 +24,26 @@ from utils.misc import calculate_sha256
 from typing import Optional
 from pydantic import BaseModel
 from pathlib import Path
+import mimetypes
 import uuid
 import base64
 import json
 import logging
 
-from config import SRC_LOG_LEVELS, CACHE_DIR, AUTOMATIC1111_BASE_URL, COMFYUI_BASE_URL
+from config import (
+    SRC_LOG_LEVELS,
+    CACHE_DIR,
+    IMAGE_GENERATION_ENGINE,
+    ENABLE_IMAGE_GENERATION,
+    AUTOMATIC1111_BASE_URL,
+    COMFYUI_BASE_URL,
+    IMAGES_OPENAI_API_BASE_URL,
+    IMAGES_OPENAI_API_KEY,
+    IMAGE_GENERATION_MODEL,
+    IMAGE_SIZE,
+    IMAGE_STEPS,
+    AppConfig,
+)
 
 
 log = logging.getLogger(__name__)
@@ -47,24 +61,31 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
-app.state.ENGINE = ""
-app.state.ENABLED = False
+app.state.config = AppConfig()
 
-app.state.OPENAI_API_KEY = ""
-app.state.MODEL = ""
+app.state.config.ENGINE = IMAGE_GENERATION_ENGINE
+app.state.config.ENABLED = ENABLE_IMAGE_GENERATION
+
+app.state.config.OPENAI_API_BASE_URL = IMAGES_OPENAI_API_BASE_URL
+app.state.config.OPENAI_API_KEY = IMAGES_OPENAI_API_KEY
+
+app.state.config.MODEL = IMAGE_GENERATION_MODEL
 
 
-app.state.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
-app.state.COMFYUI_BASE_URL = COMFYUI_BASE_URL
+app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
+app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL
 
 
-app.state.IMAGE_SIZE = "512x512"
-app.state.IMAGE_STEPS = 50
+app.state.config.IMAGE_SIZE = IMAGE_SIZE
+app.state.config.IMAGE_STEPS = IMAGE_STEPS
 
 
 @app.get("/config")
 async def get_config(request: Request, user=Depends(get_admin_user)):
-    return {"engine": app.state.ENGINE, "enabled": app.state.ENABLED}
+    return {
+        "engine": app.state.config.ENGINE,
+        "enabled": app.state.config.ENABLED,
+    }
 
 
 class ConfigUpdateForm(BaseModel):
@@ -74,9 +95,12 @@ class ConfigUpdateForm(BaseModel):
 
 @app.post("/config/update")
 async def update_config(form_data: ConfigUpdateForm, user=Depends(get_admin_user)):
-    app.state.ENGINE = form_data.engine
-    app.state.ENABLED = form_data.enabled
-    return {"engine": app.state.ENGINE, "enabled": app.state.ENABLED}
+    app.state.config.ENGINE = form_data.engine
+    app.state.config.ENABLED = form_data.enabled
+    return {
+        "engine": app.state.config.ENGINE,
+        "enabled": app.state.config.ENABLED,
+    }
 
 
 class EngineUrlUpdateForm(BaseModel):
@@ -87,8 +111,8 @@ class EngineUrlUpdateForm(BaseModel):
 @app.get("/url")
 async def get_engine_url(user=Depends(get_admin_user)):
     return {
-        "AUTOMATIC1111_BASE_URL": app.state.AUTOMATIC1111_BASE_URL,
-        "COMFYUI_BASE_URL": app.state.COMFYUI_BASE_URL,
+        "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
+        "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
     }
 
 
@@ -98,54 +122,60 @@ async def update_engine_url(
 ):
 
     if form_data.AUTOMATIC1111_BASE_URL == None:
-        app.state.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
+        app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
     else:
         url = form_data.AUTOMATIC1111_BASE_URL.strip("/")
         try:
             r = requests.head(url)
-            app.state.AUTOMATIC1111_BASE_URL = url
+            app.state.config.AUTOMATIC1111_BASE_URL = url
         except Exception as e:
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
     if form_data.COMFYUI_BASE_URL == None:
-        app.state.COMFYUI_BASE_URL = COMFYUI_BASE_URL
+        app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL
     else:
         url = form_data.COMFYUI_BASE_URL.strip("/")
 
         try:
             r = requests.head(url)
-            app.state.COMFYUI_BASE_URL = url
+            app.state.config.COMFYUI_BASE_URL = url
         except Exception as e:
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
     return {
-        "AUTOMATIC1111_BASE_URL": app.state.AUTOMATIC1111_BASE_URL,
-        "COMFYUI_BASE_URL": app.state.COMFYUI_BASE_URL,
+        "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
+        "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
         "status": True,
     }
 
 
-class OpenAIKeyUpdateForm(BaseModel):
+class OpenAIConfigUpdateForm(BaseModel):
+    url: str
     key: str
 
 
-@app.get("/key")
-async def get_openai_key(user=Depends(get_admin_user)):
-    return {"OPENAI_API_KEY": app.state.OPENAI_API_KEY}
+@app.get("/openai/config")
+async def get_openai_config(user=Depends(get_admin_user)):
+    return {
+        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+    }
 
 
-@app.post("/key/update")
-async def update_openai_key(
-    form_data: OpenAIKeyUpdateForm, user=Depends(get_admin_user)
+@app.post("/openai/config/update")
+async def update_openai_config(
+    form_data: OpenAIConfigUpdateForm, user=Depends(get_admin_user)
 ):
-
     if form_data.key == "":
         raise HTTPException(status_code=400, detail=ERROR_MESSAGES.API_KEY_NOT_FOUND)
 
-    app.state.OPENAI_API_KEY = form_data.key
+    app.state.config.OPENAI_API_BASE_URL = form_data.url
+    app.state.config.OPENAI_API_KEY = form_data.key
+
     return {
-        "OPENAI_API_KEY": app.state.OPENAI_API_KEY,
         "status": True,
+        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
     }
 
 
@@ -155,7 +185,7 @@ class ImageSizeUpdateForm(BaseModel):
 
 @app.get("/size")
 async def get_image_size(user=Depends(get_admin_user)):
-    return {"IMAGE_SIZE": app.state.IMAGE_SIZE}
+    return {"IMAGE_SIZE": app.state.config.IMAGE_SIZE}
 
 
 @app.post("/size/update")
@@ -164,9 +194,9 @@ async def update_image_size(
 ):
     pattern = r"^\d+x\d+$"  # Regular expression pattern
     if re.match(pattern, form_data.size):
-        app.state.IMAGE_SIZE = form_data.size
+        app.state.config.IMAGE_SIZE = form_data.size
         return {
-            "IMAGE_SIZE": app.state.IMAGE_SIZE,
+            "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
             "status": True,
         }
     else:
@@ -182,7 +212,7 @@ class ImageStepsUpdateForm(BaseModel):
 
 @app.get("/steps")
 async def get_image_size(user=Depends(get_admin_user)):
-    return {"IMAGE_STEPS": app.state.IMAGE_STEPS}
+    return {"IMAGE_STEPS": app.state.config.IMAGE_STEPS}
 
 
 @app.post("/steps/update")
@@ -190,9 +220,9 @@ async def update_image_size(
     form_data: ImageStepsUpdateForm, user=Depends(get_admin_user)
 ):
     if form_data.steps >= 0:
-        app.state.IMAGE_STEPS = form_data.steps
+        app.state.config.IMAGE_STEPS = form_data.steps
         return {
-            "IMAGE_STEPS": app.state.IMAGE_STEPS,
+            "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
             "status": True,
         }
     else:
@@ -205,14 +235,14 @@ async def update_image_size(
 @app.get("/models")
 def get_models(user=Depends(get_current_user)):
     try:
-        if app.state.ENGINE == "openai":
+        if app.state.config.ENGINE == "openai":
             return [
                 {"id": "dall-e-2", "name": "DALL·E 2"},
                 {"id": "dall-e-3", "name": "DALL·E 3"},
             ]
-        elif app.state.ENGINE == "comfyui":
+        elif app.state.config.ENGINE == "comfyui":
 
-            r = requests.get(url=f"{app.state.COMFYUI_BASE_URL}/object_info")
+            r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info")
             info = r.json()
 
             return list(
@@ -224,7 +254,7 @@ def get_models(user=Depends(get_current_user)):
 
         else:
             r = requests.get(
-                url=f"{app.state.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models"
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models"
             )
             models = r.json()
             return list(
@@ -234,23 +264,29 @@ def get_models(user=Depends(get_current_user)):
                 )
             )
     except Exception as e:
-        app.state.ENABLED = False
+        app.state.config.ENABLED = False
         raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
 
 @app.get("/models/default")
 async def get_default_model(user=Depends(get_admin_user)):
     try:
-        if app.state.ENGINE == "openai":
-            return {"model": app.state.MODEL if app.state.MODEL else "dall-e-2"}
-        elif app.state.ENGINE == "comfyui":
-            return {"model": app.state.MODEL if app.state.MODEL else ""}
+        if app.state.config.ENGINE == "openai":
+            return {
+                "model": (
+                    app.state.config.MODEL if app.state.config.MODEL else "dall-e-2"
+                )
+            }
+        elif app.state.config.ENGINE == "comfyui":
+            return {"model": (app.state.config.MODEL if app.state.config.MODEL else "")}
         else:
-            r = requests.get(url=f"{app.state.AUTOMATIC1111_BASE_URL}/sdapi/v1/options")
+            r = requests.get(
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options"
+            )
             options = r.json()
             return {"model": options["sd_model_checkpoint"]}
     except Exception as e:
-        app.state.ENABLED = False
+        app.state.config.ENABLED = False
         raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
 
@@ -259,20 +295,20 @@ class UpdateModelForm(BaseModel):
 
 
 def set_model_handler(model: str):
-    if app.state.ENGINE == "openai":
-        app.state.MODEL = model
-        return app.state.MODEL
-    if app.state.ENGINE == "comfyui":
-        app.state.MODEL = model
-        return app.state.MODEL
+    if app.state.config.ENGINE in ["openai", "comfyui"]:
+        app.state.config.MODEL = model
+        return app.state.config.MODEL
     else:
-        r = requests.get(url=f"{app.state.AUTOMATIC1111_BASE_URL}/sdapi/v1/options")
+        r = requests.get(
+            url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options"
+        )
         options = r.json()
 
         if model != options["sd_model_checkpoint"]:
             options["sd_model_checkpoint"] = model
             r = requests.post(
-                url=f"{app.state.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", json=options
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                json=options,
             )
 
         return options
@@ -295,35 +331,61 @@ class GenerateImageForm(BaseModel):
 
 
 def save_b64_image(b64_str):
-    image_id = str(uuid.uuid4())
-    file_path = IMAGE_CACHE_DIR.joinpath(f"{image_id}.png")
-
     try:
-        # Split the base64 string to get the actual image data
-        img_data = base64.b64decode(b64_str)
+        image_id = str(uuid.uuid4())
 
-        # Write the image data to a file
-        with open(file_path, "wb") as f:
-            f.write(img_data)
+        if "," in b64_str:
+            header, encoded = b64_str.split(",", 1)
+            mime_type = header.split(";")[0]
+
+            img_data = base64.b64decode(encoded)
+            image_format = mimetypes.guess_extension(mime_type)
+
+            image_filename = f"{image_id}{image_format}"
+            file_path = IMAGE_CACHE_DIR / f"{image_filename}"
+            with open(file_path, "wb") as f:
+                f.write(img_data)
+            return image_filename
+        else:
+            image_filename = f"{image_id}.png"
+            file_path = IMAGE_CACHE_DIR.joinpath(image_filename)
+
+            img_data = base64.b64decode(b64_str)
+
+            # Write the image data to a file
+            with open(file_path, "wb") as f:
+                f.write(img_data)
+            return image_filename
 
-        return image_id
     except Exception as e:
-        log.error(f"Error saving image: {e}")
+        log.exception(f"Error saving image: {e}")
         return None
 
 
 def save_url_image(url):
     image_id = str(uuid.uuid4())
-    file_path = IMAGE_CACHE_DIR.joinpath(f"{image_id}.png")
-
     try:
         r = requests.get(url)
         r.raise_for_status()
+        if r.headers["content-type"].split("/")[0] == "image":
 
-        with open(file_path, "wb") as image_file:
-            image_file.write(r.content)
+            mime_type = r.headers["content-type"]
+            image_format = mimetypes.guess_extension(mime_type)
+
+            if not image_format:
+                raise ValueError("Could not determine image type from MIME type")
+
+            image_filename = f"{image_id}{image_format}"
+
+            file_path = IMAGE_CACHE_DIR.joinpath(f"{image_filename}")
+            with open(file_path, "wb") as image_file:
+                for chunk in r.iter_content(chunk_size=8192):
+                    image_file.write(chunk)
+            return image_filename
+        else:
+            log.error(f"Url does not point to an image.")
+            return None
 
-        return image_id
     except Exception as e:
         log.exception(f"Error saving image: {e}")
         return None
@@ -335,26 +397,32 @@ def generate_image(
     user=Depends(get_current_user),
 ):
 
-    width, height = tuple(map(int, app.state.IMAGE_SIZE.split("x")))
+    width, height = tuple(map(int, app.state.config.IMAGE_SIZE.split("x")))
 
     r = None
     try:
-        if app.state.ENGINE == "openai":
+        if app.state.config.ENGINE == "openai":
 
             headers = {}
-            headers["Authorization"] = f"Bearer {app.state.OPENAI_API_KEY}"
+            headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}"
             headers["Content-Type"] = "application/json"
 
             data = {
-                "model": app.state.MODEL if app.state.MODEL != "" else "dall-e-2",
+                "model": (
+                    app.state.config.MODEL
+                    if app.state.config.MODEL != ""
+                    else "dall-e-2"
+                ),
                 "prompt": form_data.prompt,
                 "n": form_data.n,
-                "size": form_data.size if form_data.size else app.state.IMAGE_SIZE,
+                "size": (
+                    form_data.size if form_data.size else app.state.config.IMAGE_SIZE
+                ),
                 "response_format": "b64_json",
             }
 
             r = requests.post(
-                url=f"https://api.openai.com/v1/images/generations",
+                url=f"{app.state.config.OPENAI_API_BASE_URL}/images/generations",
                 json=data,
                 headers=headers,
             )
@@ -365,16 +433,16 @@ def generate_image(
             images = []
 
             for image in res["data"]:
-                image_id = save_b64_image(image["b64_json"])
-                images.append({"url": f"/cache/image/generations/{image_id}.png"})
-                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_id}.json")
+                image_filename = save_b64_image(image["b64_json"])
+                images.append({"url": f"/cache/image/generations/{image_filename}"})
+                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_filename}.json")
 
                 with open(file_body_path, "w") as f:
                     json.dump(data, f)
 
             return images
 
-        elif app.state.ENGINE == "comfyui":
+        elif app.state.config.ENGINE == "comfyui":
 
             data = {
                 "prompt": form_data.prompt,
@@ -383,28 +451,28 @@ def generate_image(
                 "n": form_data.n,
             }
 
-            if app.state.IMAGE_STEPS != None:
-                data["steps"] = app.state.IMAGE_STEPS
+            if app.state.config.IMAGE_STEPS is not None:
+                data["steps"] = app.state.config.IMAGE_STEPS
 
-            if form_data.negative_prompt != None:
+            if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
 
             data = ImageGenerationPayload(**data)
 
             res = comfyui_generate_image(
-                app.state.MODEL,
+                app.state.config.MODEL,
                 data,
                 user.id,
-                app.state.COMFYUI_BASE_URL,
+                app.state.config.COMFYUI_BASE_URL,
             )
             log.debug(f"res: {res}")
 
             images = []
 
             for image in res["data"]:
-                image_id = save_url_image(image["url"])
-                images.append({"url": f"/cache/image/generations/{image_id}.png"})
-                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_id}.json")
+                image_filename = save_url_image(image["url"])
+                images.append({"url": f"/cache/image/generations/{image_filename}"})
+                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_filename}.json")
 
                 with open(file_body_path, "w") as f:
                     json.dump(data.model_dump(exclude_none=True), f)
@@ -422,14 +490,14 @@ def generate_image(
                 "height": height,
             }
 
-            if app.state.IMAGE_STEPS != None:
-                data["steps"] = app.state.IMAGE_STEPS
+            if app.state.config.IMAGE_STEPS is not None:
+                data["steps"] = app.state.config.IMAGE_STEPS
 
-            if form_data.negative_prompt != None:
+            if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
 
             r = requests.post(
-                url=f"{app.state.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img",
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img",
                 json=data,
             )
 
@@ -440,9 +508,9 @@ def generate_image(
             images = []
 
             for image in res["images"]:
-                image_id = save_b64_image(image)
-                images.append({"url": f"/cache/image/generations/{image_id}.png"})
-                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_id}.json")
+                image_filename = save_b64_image(image)
+                images.append({"url": f"/cache/image/generations/{image_filename}"})
+                file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_filename}.json")
 
                 with open(file_body_path, "w") as f:
                     json.dump({**data, "info": res["info"]}, f)

backend/apps/images/utils/comfyui.py

@@ -195,7 +195,7 @@ class ImageGenerationPayload(BaseModel):
 def comfyui_generate_image(
     model: str, payload: ImageGenerationPayload, client_id, base_url
 ):
-    host = base_url.replace("http://", "").replace("https://", "")
+    ws_url = base_url.replace("http://", "ws://").replace("https://", "wss://")
 
     comfyui_prompt = json.loads(COMFYUI_DEFAULT_PROMPT)
 
@@ -217,7 +217,7 @@ def comfyui_generate_image(
 
     try:
         ws = websocket.WebSocket()
-        ws.connect(f"ws://{host}/ws?clientId={client_id}")
+        ws.connect(f"{ws_url}/ws?clientId={client_id}")
         log.info("WebSocket connection established.")
     except Exception as e:
         log.exception(f"Failed to connect to WebSocket server: {e}")

backend/apps/litellm/main.py

@@ -1,100 +1,379 @@
+import sys
+from contextlib import asynccontextmanager
+
+from fastapi import FastAPI, Depends, HTTPException
+from fastapi.routing import APIRoute
+from fastapi.middleware.cors import CORSMiddleware
+
 import logging
-
-from litellm.proxy.proxy_server import ProxyConfig, initialize
-from litellm.proxy.proxy_server import app
-
 from fastapi import FastAPI, Request, Depends, status, Response
 from fastapi.responses import JSONResponse
 
 from starlette.middleware.base import BaseHTTPMiddleware, RequestResponseEndpoint
 from starlette.responses import StreamingResponse
 import json
+import time
+import requests
 
-from utils.utils import get_http_authorization_cred, get_current_user
+from pydantic import BaseModel, ConfigDict
+from typing import Optional, List
+
+from utils.utils import get_verified_user, get_current_user, get_admin_user
 from config import SRC_LOG_LEVELS, ENV
+from constants import MESSAGES
+
+import os
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["LITELLM"])
 
 
 from config import (
-    MODEL_FILTER_ENABLED,
+    ENABLE_LITELLM,
+    ENABLE_MODEL_FILTER,
     MODEL_FILTER_LIST,
+    DATA_DIR,
+    LITELLM_PROXY_PORT,
+    LITELLM_PROXY_HOST,
+)
+
+import warnings
+
+warnings.simplefilter("ignore")
+
+from litellm.utils import get_llm_provider
+
+import asyncio
+import subprocess
+import yaml
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    log.info("startup_event")
+    # TODO: Check config.yaml file and create one
+    asyncio.create_task(start_litellm_background())
+    yield
+
+
+app = FastAPI(lifespan=lifespan)
+
+origins = ["*"]
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=origins,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
 )
 
 
-proxy_config = ProxyConfig()
+LITELLM_CONFIG_DIR = f"{DATA_DIR}/litellm/config.yaml"
+
+with open(LITELLM_CONFIG_DIR, "r") as file:
+    litellm_config = yaml.safe_load(file)
 
 
-async def config():
-    router, model_list, general_settings = await proxy_config.load_config(
-        router=None, config_file_path="./data/litellm/config.yaml"
-    )
-
-    await initialize(config="./data/litellm/config.yaml", telemetry=False)
+app.state.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER.value
+app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST.value
 
 
-async def startup():
-    await config()
+app.state.ENABLE = ENABLE_LITELLM
+app.state.CONFIG = litellm_config
+
+# Global variable to store the subprocess reference
+background_process = None
+
+CONFLICT_ENV_VARS = [
+    # Uvicorn uses PORT, so LiteLLM might use it as well
+    "PORT",
+    # LiteLLM uses DATABASE_URL for Prisma connections
+    "DATABASE_URL",
+]
 
 
-@app.on_event("startup")
-async def on_startup():
-    await startup()
-
-
-app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
-app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
-
-
-@app.middleware("http")
-async def auth_middleware(request: Request, call_next):
-    auth_header = request.headers.get("Authorization", "")
-    request.state.user = None
+async def run_background_process(command):
+    global background_process
+    log.info("run_background_process")
 
     try:
-        user = get_current_user(get_http_authorization_cred(auth_header))
-        log.debug(f"user: {user}")
-        request.state.user = user
+        # Log the command to be executed
+        log.info(f"Executing command: {command}")
+        # Filter environment variables known to conflict with litellm
+        env = {k: v for k, v in os.environ.items() if k not in CONFLICT_ENV_VARS}
+        # Execute the command and create a subprocess
+        process = await asyncio.create_subprocess_exec(
+            *command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, env=env
+        )
+        background_process = process
+        log.info("Subprocess started successfully.")
+
+        # Capture STDERR for debugging purposes
+        stderr_output = await process.stderr.read()
+        stderr_text = stderr_output.decode().strip()
+        if stderr_text:
+            log.info(f"Subprocess STDERR: {stderr_text}")
+
+        # log.info output line by line
+        async for line in process.stdout:
+            log.info(line.decode().strip())
+
+        # Wait for the process to finish
+        returncode = await process.wait()
+        log.info(f"Subprocess exited with return code {returncode}")
     except Exception as e:
-        return JSONResponse(status_code=400, content={"detail": str(e)})
-
-    response = await call_next(request)
-    return response
+        log.error(f"Failed to start subprocess: {e}")
+        raise  # Optionally re-raise the exception if you want it to propagate
 
 
-class ModifyModelsResponseMiddleware(BaseHTTPMiddleware):
-    async def dispatch(
-        self, request: Request, call_next: RequestResponseEndpoint
-    ) -> Response:
+async def start_litellm_background():
+    log.info("start_litellm_background")
+    # Command to run in the background
+    command = [
+        "litellm",
+        "--port",
+        str(LITELLM_PROXY_PORT),
+        "--host",
+        LITELLM_PROXY_HOST,
+        "--telemetry",
+        "False",
+        "--config",
+        LITELLM_CONFIG_DIR,
+    ]
 
-        response = await call_next(request)
-        user = request.state.user
+    await run_background_process(command)
 
-        if "/models" in request.url.path:
-            if isinstance(response, StreamingResponse):
-                # Read the content of the streaming response
-                body = b""
-                async for chunk in response.body_iterator:
-                    body += chunk
 
-                data = json.loads(body.decode("utf-8"))
+async def shutdown_litellm_background():
+    log.info("shutdown_litellm_background")
+    global background_process
+    if background_process:
+        background_process.terminate()
+        await background_process.wait()  # Ensure the process has terminated
+        log.info("Subprocess terminated")
+        background_process = None
 
-                if app.state.MODEL_FILTER_ENABLED:
-                    if user and user.role == "user":
-                        data["data"] = list(
-                            filter(
-                                lambda model: model["id"]
-                                in app.state.MODEL_FILTER_LIST,
-                                data["data"],
-                            )
+
+@app.get("/")
+async def get_status():
+    return {"status": True}
+
+
+async def restart_litellm():
+    """
+    Endpoint to restart the litellm background service.
+    """
+    log.info("Requested restart of litellm service.")
+    try:
+        # Shut down the existing process if it is running
+        await shutdown_litellm_background()
+        log.info("litellm service shutdown complete.")
+
+        # Restart the background service
+
+        asyncio.create_task(start_litellm_background())
+        log.info("litellm service restart complete.")
+
+        return {
+            "status": "success",
+            "message": "litellm service restarted successfully.",
+        }
+    except Exception as e:
+        log.info(f"Error restarting litellm service: {e}")
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, detail=str(e)
+        )
+
+
+@app.get("/restart")
+async def restart_litellm_handler(user=Depends(get_admin_user)):
+    return await restart_litellm()
+
+
+@app.get("/config")
+async def get_config(user=Depends(get_admin_user)):
+    return app.state.CONFIG
+
+
+class LiteLLMConfigForm(BaseModel):
+    general_settings: Optional[dict] = None
+    litellm_settings: Optional[dict] = None
+    model_list: Optional[List[dict]] = None
+    router_settings: Optional[dict] = None
+
+    model_config = ConfigDict(protected_namespaces=())
+
+
+@app.post("/config/update")
+async def update_config(form_data: LiteLLMConfigForm, user=Depends(get_admin_user)):
+    app.state.CONFIG = form_data.model_dump(exclude_none=True)
+
+    with open(LITELLM_CONFIG_DIR, "w") as file:
+        yaml.dump(app.state.CONFIG, file)
+
+    await restart_litellm()
+    return app.state.CONFIG
+
+
+@app.get("/models")
+@app.get("/v1/models")
+async def get_models(user=Depends(get_current_user)):
+
+    if app.state.ENABLE:
+        while not background_process:
+            await asyncio.sleep(0.1)
+
+        url = f"http://localhost:{LITELLM_PROXY_PORT}/v1"
+        r = None
+        try:
+            r = requests.request(method="GET", url=f"{url}/models")
+            r.raise_for_status()
+
+            data = r.json()
+
+            if app.state.ENABLE_MODEL_FILTER:
+                if user and user.role == "user":
+                    data["data"] = list(
+                        filter(
+                            lambda model: model["id"] in app.state.MODEL_FILTER_LIST,
+                            data["data"],
                         )
+                    )
 
-                # Modified Flag
-                data["modified"] = True
-                return JSONResponse(content=data)
+            return data
+        except Exception as e:
 
-        return response
+            log.exception(e)
+            error_detail = "Open WebUI: Server Connection Error"
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "error" in res:
+                        error_detail = f"External: {res['error']}"
+                except:
+                    error_detail = f"External: {e}"
+
+            return {
+                "data": [
+                    {
+                        "id": model["model_name"],
+                        "object": "model",
+                        "created": int(time.time()),
+                        "owned_by": "openai",
+                    }
+                    for model in app.state.CONFIG["model_list"]
+                ],
+                "object": "list",
+            }
+    else:
+        return {
+            "data": [],
+            "object": "list",
+        }
 
 
-app.add_middleware(ModifyModelsResponseMiddleware)
+@app.get("/model/info")
+async def get_model_list(user=Depends(get_admin_user)):
+    return {"data": app.state.CONFIG["model_list"]}
+
+
+class AddLiteLLMModelForm(BaseModel):
+    model_name: str
+    litellm_params: dict
+
+    model_config = ConfigDict(protected_namespaces=())
+
+
+@app.post("/model/new")
+async def add_model_to_config(
+    form_data: AddLiteLLMModelForm, user=Depends(get_admin_user)
+):
+    try:
+        get_llm_provider(model=form_data.model_name)
+        app.state.CONFIG["model_list"].append(form_data.model_dump())
+
+        with open(LITELLM_CONFIG_DIR, "w") as file:
+            yaml.dump(app.state.CONFIG, file)
+
+        await restart_litellm()
+
+        return {"message": MESSAGES.MODEL_ADDED(form_data.model_name)}
+    except Exception as e:
+        print(e)
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, detail=str(e)
+        )
+
+
+class DeleteLiteLLMModelForm(BaseModel):
+    id: str
+
+
+@app.post("/model/delete")
+async def delete_model_from_config(
+    form_data: DeleteLiteLLMModelForm, user=Depends(get_admin_user)
+):
+    app.state.CONFIG["model_list"] = [
+        model
+        for model in app.state.CONFIG["model_list"]
+        if model["model_name"] != form_data.id
+    ]
+
+    with open(LITELLM_CONFIG_DIR, "w") as file:
+        yaml.dump(app.state.CONFIG, file)
+
+    await restart_litellm()
+
+    return {"message": MESSAGES.MODEL_DELETED(form_data.id)}
+
+
+@app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
+async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
+    body = await request.body()
+
+    url = f"http://localhost:{LITELLM_PROXY_PORT}"
+
+    target_url = f"{url}/{path}"
+
+    headers = {}
+    # headers["Authorization"] = f"Bearer {key}"
+    headers["Content-Type"] = "application/json"
+
+    r = None
+
+    try:
+        r = requests.request(
+            method=request.method,
+            url=target_url,
+            data=body,
+            headers=headers,
+            stream=True,
+        )
+
+        r.raise_for_status()
+
+        # Check if response is SSE
+        if "text/event-stream" in r.headers.get("Content-Type", ""):
+            return StreamingResponse(
+                r.iter_content(chunk_size=8192),
+                status_code=r.status_code,
+                headers=dict(r.headers),
+            )
+        else:
+            response_data = r.json()
+            return response_data
+    except Exception as e:
+        log.exception(e)
+        error_detail = "Open WebUI: Server Connection Error"
+        if r is not None:
+            try:
+                res = r.json()
+                if "error" in res:
+                    error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
+            except:
+                error_detail = f"External: {e}"
+
+        raise HTTPException(
+            status_code=r.status_code if r else 500, detail=error_detail
+        )

backend/apps/ollama/main.py

@@ -16,6 +16,7 @@ from fastapi.concurrency import run_in_threadpool
 from pydantic import BaseModel, ConfigDict
 
 import os
+import re
 import copy
 import random
 import requests
@@ -24,21 +25,28 @@ import uuid
 import aiohttp
 import asyncio
 import logging
+import time
 from urllib.parse import urlparse
 from typing import Optional, List, Union
 
 
 from apps.web.models.users import Users
 from constants import ERROR_MESSAGES
-from utils.utils import decode_token, get_current_user, get_admin_user
+from utils.utils import (
+    decode_token,
+    get_current_user,
+    get_verified_user,
+    get_admin_user,
+)
 
 
 from config import (
     SRC_LOG_LEVELS,
     OLLAMA_BASE_URLS,
-    MODEL_FILTER_ENABLED,
+    ENABLE_MODEL_FILTER,
     MODEL_FILTER_LIST,
     UPLOAD_DIR,
+    AppConfig,
 )
 from utils.misc import calculate_sha256
 
@@ -54,11 +62,12 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+app.state.config = AppConfig()
 
-app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
-app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+app.state.config.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
+app.state.config.MODEL_FILTER_LIST = MODEL_FILTER_LIST
 
-app.state.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS
+app.state.config.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS
 app.state.MODELS = {}
 
 
@@ -89,7 +98,7 @@ async def get_status():
 
 @app.get("/urls")
 async def get_ollama_api_urls(user=Depends(get_admin_user)):
-    return {"OLLAMA_BASE_URLS": app.state.OLLAMA_BASE_URLS}
+    return {"OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS}
 
 
 class UrlUpdateForm(BaseModel):
@@ -98,10 +107,10 @@ class UrlUpdateForm(BaseModel):
 
 @app.post("/urls/update")
 async def update_ollama_api_url(form_data: UrlUpdateForm, user=Depends(get_admin_user)):
-    app.state.OLLAMA_BASE_URLS = form_data.urls
+    app.state.config.OLLAMA_BASE_URLS = form_data.urls
 
-    log.info(f"app.state.OLLAMA_BASE_URLS: {app.state.OLLAMA_BASE_URLS}")
-    return {"OLLAMA_BASE_URLS": app.state.OLLAMA_BASE_URLS}
+    log.info(f"app.state.config.OLLAMA_BASE_URLS: {app.state.config.OLLAMA_BASE_URLS}")
+    return {"OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS}
 
 
 @app.get("/cancel/{request_id}")
@@ -115,8 +124,9 @@ async def cancel_ollama_request(request_id: str, user=Depends(get_current_user))
 
 
 async def fetch_url(url):
+    timeout = aiohttp.ClientTimeout(total=5)
     try:
-        async with aiohttp.ClientSession() as session:
+        async with aiohttp.ClientSession(timeout=timeout) as session:
             async with session.get(url) as response:
                 return await response.json()
     except Exception as e:
@@ -146,7 +156,7 @@ def merge_models_lists(model_lists):
 
 async def get_all_models():
     log.info("get_all_models()")
-    tasks = [fetch_url(f"{url}/api/tags") for url in app.state.OLLAMA_BASE_URLS]
+    tasks = [fetch_url(f"{url}/api/tags") for url in app.state.config.OLLAMA_BASE_URLS]
     responses = await asyncio.gather(*tasks)
 
     models = {
@@ -163,23 +173,24 @@ async def get_all_models():
 @app.get("/api/tags")
 @app.get("/api/tags/{url_idx}")
 async def get_ollama_tags(
-    url_idx: Optional[int] = None, user=Depends(get_current_user)
+    url_idx: Optional[int] = None, user=Depends(get_verified_user)
 ):
     if url_idx == None:
         models = await get_all_models()
 
-        if app.state.MODEL_FILTER_ENABLED:
+        if app.state.config.ENABLE_MODEL_FILTER:
             if user.role == "user":
                 models["models"] = list(
                     filter(
-                        lambda model: model["name"] in app.state.MODEL_FILTER_LIST,
+                        lambda model: model["name"]
+                        in app.state.config.MODEL_FILTER_LIST,
                         models["models"],
                     )
                 )
                 return models
         return models
     else:
-        url = app.state.OLLAMA_BASE_URLS[url_idx]
+        url = app.state.config.OLLAMA_BASE_URLS[url_idx]
         try:
             r = requests.request(method="GET", url=f"{url}/api/tags")
             r.raise_for_status()
@@ -209,14 +220,18 @@ async def get_ollama_versions(url_idx: Optional[int] = None):
     if url_idx == None:
 
         # returns lowest version
-        tasks = [fetch_url(f"{url}/api/version") for url in app.state.OLLAMA_BASE_URLS]
+        tasks = [
+            fetch_url(f"{url}/api/version") for url in app.state.config.OLLAMA_BASE_URLS
+        ]
         responses = await asyncio.gather(*tasks)
         responses = list(filter(lambda x: x is not None, responses))
 
         if len(responses) > 0:
             lowest_version = min(
                 responses,
-                key=lambda x: tuple(map(int, x["version"].split("-")[0].split("."))),
+                key=lambda x: tuple(
+                    map(int, re.sub(r"^v|-.*", "", x["version"]).split("."))
+                ),
             )
 
             return {"version": lowest_version["version"]}
@@ -226,7 +241,7 @@ async def get_ollama_versions(url_idx: Optional[int] = None):
                 detail=ERROR_MESSAGES.OLLAMA_NOT_FOUND,
             )
     else:
-        url = app.state.OLLAMA_BASE_URLS[url_idx]
+        url = app.state.config.OLLAMA_BASE_URLS[url_idx]
         try:
             r = requests.request(method="GET", url=f"{url}/api/version")
             r.raise_for_status()
@@ -258,7 +273,7 @@ class ModelNameForm(BaseModel):
 async def pull_model(
     form_data: ModelNameForm, url_idx: int = 0, user=Depends(get_admin_user)
 ):
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     r = None
@@ -346,7 +361,7 @@ async def push_model(
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.debug(f"url: {url}")
 
     r = None
@@ -408,7 +423,7 @@ async def create_model(
     form_data: CreateModelForm, url_idx: int = 0, user=Depends(get_admin_user)
 ):
     log.debug(f"form_data: {form_data}")
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     r = None
@@ -481,7 +496,7 @@ async def copy_model(
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.source),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     try:
@@ -528,7 +543,7 @@ async def delete_model(
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     try:
@@ -560,7 +575,7 @@ async def delete_model(
 
 
 @app.post("/api/show")
-async def show_model_info(form_data: ModelNameForm, user=Depends(get_current_user)):
+async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_user)):
     if form_data.name not in app.state.MODELS:
         raise HTTPException(
             status_code=400,
@@ -568,7 +583,7 @@ async def show_model_info(form_data: ModelNameForm, user=Depends(get_current_use
         )
 
     url_idx = random.choice(app.state.MODELS[form_data.name]["urls"])
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     try:
@@ -609,18 +624,23 @@ class GenerateEmbeddingsForm(BaseModel):
 async def generate_embeddings(
     form_data: GenerateEmbeddingsForm,
     url_idx: Optional[int] = None,
-    user=Depends(get_current_user),
+    user=Depends(get_verified_user),
 ):
     if url_idx == None:
-        if form_data.model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[form_data.model]["urls"])
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     try:
@@ -649,6 +669,60 @@ async def generate_embeddings(
         )
 
 
+def generate_ollama_embeddings(
+    form_data: GenerateEmbeddingsForm,
+    url_idx: Optional[int] = None,
+):
+
+    log.info(f"generate_ollama_embeddings {form_data}")
+
+    if url_idx == None:
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
+        else:
+            raise HTTPException(
+                status_code=400,
+                detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
+            )
+
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    log.info(f"url: {url}")
+
+    try:
+        r = requests.request(
+            method="POST",
+            url=f"{url}/api/embeddings",
+            data=form_data.model_dump_json(exclude_none=True).encode(),
+        )
+        r.raise_for_status()
+
+        data = r.json()
+
+        log.info(f"generate_ollama_embeddings {data}")
+
+        if "embedding" in data:
+            return data["embedding"]
+        else:
+            raise "Something went wrong :/"
+    except Exception as e:
+        log.exception(e)
+        error_detail = "Open WebUI: Server Connection Error"
+        if r is not None:
+            try:
+                res = r.json()
+                if "error" in res:
+                    error_detail = f"Ollama: {res['error']}"
+            except:
+                error_detail = f"Ollama: {e}"
+
+        raise error_detail
+
+
 class GenerateCompletionForm(BaseModel):
     model: str
     prompt: str
@@ -668,19 +742,24 @@ class GenerateCompletionForm(BaseModel):
 async def generate_completion(
     form_data: GenerateCompletionForm,
     url_idx: Optional[int] = None,
-    user=Depends(get_current_user),
+    user=Depends(get_verified_user),
 ):
 
     if url_idx == None:
-        if form_data.model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[form_data.model]["urls"])
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     r = None
@@ -766,19 +845,24 @@ class GenerateChatCompletionForm(BaseModel):
 async def generate_chat_completion(
     form_data: GenerateChatCompletionForm,
     url_idx: Optional[int] = None,
-    user=Depends(get_current_user),
+    user=Depends(get_verified_user),
 ):
 
     if url_idx == None:
-        if form_data.model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[form_data.model]["urls"])
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     r = None
@@ -870,19 +954,24 @@ class OpenAIChatCompletionForm(BaseModel):
 async def generate_openai_chat_completion(
     form_data: OpenAIChatCompletionForm,
     url_idx: Optional[int] = None,
-    user=Depends(get_current_user),
+    user=Depends(get_verified_user),
 ):
 
     if url_idx == None:
-        if form_data.model in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[form_data.model]["urls"])
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
         else:
             raise HTTPException(
                 status_code=400,
                 detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
             )
 
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
     r = None
@@ -949,6 +1038,76 @@ async def generate_openai_chat_completion(
         )
 
 
+@app.get("/v1/models")
+@app.get("/v1/models/{url_idx}")
+async def get_openai_models(
+    url_idx: Optional[int] = None,
+    user=Depends(get_verified_user),
+):
+    if url_idx == None:
+        models = await get_all_models()
+
+        if app.state.config.ENABLE_MODEL_FILTER:
+            if user.role == "user":
+                models["models"] = list(
+                    filter(
+                        lambda model: model["name"]
+                        in app.state.config.MODEL_FILTER_LIST,
+                        models["models"],
+                    )
+                )
+
+        return {
+            "data": [
+                {
+                    "id": model["model"],
+                    "object": "model",
+                    "created": int(time.time()),
+                    "owned_by": "openai",
+                }
+                for model in models["models"]
+            ],
+            "object": "list",
+        }
+
+    else:
+        url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+        try:
+            r = requests.request(method="GET", url=f"{url}/api/tags")
+            r.raise_for_status()
+
+            models = r.json()
+
+            return {
+                "data": [
+                    {
+                        "id": model["model"],
+                        "object": "model",
+                        "created": int(time.time()),
+                        "owned_by": "openai",
+                    }
+                    for model in models["models"]
+                ],
+                "object": "list",
+            }
+
+        except Exception as e:
+            log.exception(e)
+            error_detail = "Open WebUI: Server Connection Error"
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "error" in res:
+                        error_detail = f"Ollama: {res['error']}"
+                except:
+                    error_detail = f"Ollama: {e}"
+
+            raise HTTPException(
+                status_code=r.status_code if r else 500,
+                detail=error_detail,
+            )
+
+
 class UrlForm(BaseModel):
     url: str
 
@@ -1046,7 +1205,7 @@ async def download_model(
 
     if url_idx == None:
         url_idx = 0
-    url = app.state.OLLAMA_BASE_URLS[url_idx]
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 
     file_name = parse_huggingface_url(form_data.url)
 
@@ -1065,7 +1224,7 @@ async def download_model(
 def upload_model(file: UploadFile = File(...), url_idx: Optional[int] = None):
     if url_idx == None:
         url_idx = 0
-    ollama_url = app.state.OLLAMA_BASE_URLS[url_idx]
+    ollama_url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 
     file_path = f"{UPLOAD_DIR}/{file.filename}"
 
@@ -1130,7 +1289,7 @@ def upload_model(file: UploadFile = File(...), url_idx: Optional[int] = None):
 # async def upload_model(file: UploadFile = File(), url_idx: Optional[int] = None):
 #     if url_idx == None:
 #         url_idx = 0
-#     url = app.state.OLLAMA_BASE_URLS[url_idx]
+#     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 
 #     file_location = os.path.join(UPLOAD_DIR, file.filename)
 #     total_size = file.size
@@ -1164,8 +1323,10 @@ def upload_model(file: UploadFile = File(...), url_idx: Optional[int] = None):
 
 
 @app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
-async def deprecated_proxy(path: str, request: Request, user=Depends(get_current_user)):
-    url = app.state.OLLAMA_BASE_URLS[0]
+async def deprecated_proxy(
+    path: str, request: Request, user=Depends(get_verified_user)
+):
+    url = app.state.config.OLLAMA_BASE_URLS[0]
     target_url = f"{url}/{path}"
 
     body = await request.body()

backend/apps/openai/main.py

@@ -21,11 +21,13 @@ from utils.utils import (
 )
 from config import (
     SRC_LOG_LEVELS,
+    ENABLE_OPENAI_API,
     OPENAI_API_BASE_URLS,
     OPENAI_API_KEYS,
     CACHE_DIR,
-    MODEL_FILTER_ENABLED,
+    ENABLE_MODEL_FILTER,
     MODEL_FILTER_LIST,
+    AppConfig,
 )
 from typing import List, Optional
 
@@ -45,11 +47,16 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
-app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
-app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
 
-app.state.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS
-app.state.OPENAI_API_KEYS = OPENAI_API_KEYS
+app.state.config = AppConfig()
+
+app.state.config.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
+app.state.config.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+
+
+app.state.config.ENABLE_OPENAI_API = ENABLE_OPENAI_API
+app.state.config.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS
+app.state.config.OPENAI_API_KEYS = OPENAI_API_KEYS
 
 app.state.MODELS = {}
 
@@ -65,6 +72,21 @@ async def check_url(request: Request, call_next):
     return response
 
 
+@app.get("/config")
+async def get_config(user=Depends(get_admin_user)):
+    return {"ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API}
+
+
+class OpenAIConfigForm(BaseModel):
+    enable_openai_api: Optional[bool] = None
+
+
+@app.post("/config/update")
+async def update_config(form_data: OpenAIConfigForm, user=Depends(get_admin_user)):
+    app.state.config.ENABLE_OPENAI_API = form_data.enable_openai_api
+    return {"ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API}
+
+
 class UrlsUpdateForm(BaseModel):
     urls: List[str]
 
@@ -75,31 +97,32 @@ class KeysUpdateForm(BaseModel):
 
 @app.get("/urls")
 async def get_openai_urls(user=Depends(get_admin_user)):
-    return {"OPENAI_API_BASE_URLS": app.state.OPENAI_API_BASE_URLS}
+    return {"OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS}
 
 
 @app.post("/urls/update")
 async def update_openai_urls(form_data: UrlsUpdateForm, user=Depends(get_admin_user)):
-    app.state.OPENAI_API_BASE_URLS = form_data.urls
-    return {"OPENAI_API_BASE_URLS": app.state.OPENAI_API_BASE_URLS}
+    await get_all_models()
+    app.state.config.OPENAI_API_BASE_URLS = form_data.urls
+    return {"OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS}
 
 
 @app.get("/keys")
 async def get_openai_keys(user=Depends(get_admin_user)):
-    return {"OPENAI_API_KEYS": app.state.OPENAI_API_KEYS}
+    return {"OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS}
 
 
 @app.post("/keys/update")
 async def update_openai_key(form_data: KeysUpdateForm, user=Depends(get_admin_user)):
-    app.state.OPENAI_API_KEYS = form_data.keys
-    return {"OPENAI_API_KEYS": app.state.OPENAI_API_KEYS}
+    app.state.config.OPENAI_API_KEYS = form_data.keys
+    return {"OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS}
 
 
 @app.post("/audio/speech")
 async def speech(request: Request, user=Depends(get_verified_user)):
     idx = None
     try:
-        idx = app.state.OPENAI_API_BASE_URLS.index("https://api.openai.com/v1")
+        idx = app.state.config.OPENAI_API_BASE_URLS.index("https://api.openai.com/v1")
         body = await request.body()
         name = hashlib.sha256(body).hexdigest()
 
@@ -113,13 +136,15 @@ async def speech(request: Request, user=Depends(get_verified_user)):
             return FileResponse(file_path)
 
         headers = {}
-        headers["Authorization"] = f"Bearer {app.state.OPENAI_API_KEYS[idx]}"
+        headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEYS[idx]}"
         headers["Content-Type"] = "application/json"
-
+        if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]:
+            headers["HTTP-Referer"] = "https://openwebui.com/"
+            headers["X-Title"] = "Open WebUI"
         r = None
         try:
             r = requests.post(
-                url=f"{app.state.OPENAI_API_BASE_URLS[idx]}/audio/speech",
+                url=f"{app.state.config.OPENAI_API_BASE_URLS[idx]}/audio/speech",
                 data=body,
                 headers=headers,
                 stream=True,
@@ -158,11 +183,15 @@ async def speech(request: Request, user=Depends(get_verified_user)):
 
 
 async def fetch_url(url, key):
+    timeout = aiohttp.ClientTimeout(total=5)
     try:
-        headers = {"Authorization": f"Bearer {key}"}
-        async with aiohttp.ClientSession() as session:
-            async with session.get(url, headers=headers) as response:
-                return await response.json()
+        if key != "":
+            headers = {"Authorization": f"Bearer {key}"}
+            async with aiohttp.ClientSession(timeout=timeout) as session:
+                async with session.get(url, headers=headers) as response:
+                    return await response.json()
+        else:
+            return None
     except Exception as e:
         # Handle connection error here
         log.error(f"Connection error: {e}")
@@ -170,6 +199,7 @@ async def fetch_url(url, key):
 
 
 def merge_models_lists(model_lists):
+    log.info(f"merge_models_lists {model_lists}")
     merged_list = []
 
     for idx, models in enumerate(model_lists):
@@ -178,7 +208,8 @@ def merge_models_lists(model_lists):
                 [
                     {**model, "urlIdx": idx}
                     for model in models
-                    if "api.openai.com" not in app.state.OPENAI_API_BASE_URLS[idx]
+                    if "api.openai.com"
+                    not in app.state.config.OPENAI_API_BASE_URLS[idx]
                     or "gpt" in model["id"]
                 ]
             )
@@ -189,23 +220,28 @@ def merge_models_lists(model_lists):
 async def get_all_models():
     log.info("get_all_models()")
 
-    if len(app.state.OPENAI_API_KEYS) == 1 and app.state.OPENAI_API_KEYS[0] == "":
+    if (
+        len(app.state.config.OPENAI_API_KEYS) == 1
+        and app.state.config.OPENAI_API_KEYS[0] == ""
+    ) or not app.state.config.ENABLE_OPENAI_API:
         models = {"data": []}
     else:
         tasks = [
-            fetch_url(f"{url}/models", app.state.OPENAI_API_KEYS[idx])
-            for idx, url in enumerate(app.state.OPENAI_API_BASE_URLS)
+            fetch_url(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx])
+            for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS)
         ]
 
         responses = await asyncio.gather(*tasks)
+        log.info(f"get_all_models:responses() {responses}")
+
         models = {
             "data": merge_models_lists(
                 list(
                     map(
                         lambda response: (
                             response["data"]
-                            if response and "data" in response
-                            else None
+                            if (response and "data" in response)
+                            else (response if isinstance(response, list) else None)
                         ),
                         responses,
                     )
@@ -224,18 +260,18 @@ async def get_all_models():
 async def get_models(url_idx: Optional[int] = None, user=Depends(get_current_user)):
     if url_idx == None:
         models = await get_all_models()
-        if app.state.MODEL_FILTER_ENABLED:
+        if app.state.config.ENABLE_MODEL_FILTER:
             if user.role == "user":
                 models["data"] = list(
                     filter(
-                        lambda model: model["id"] in app.state.MODEL_FILTER_LIST,
+                        lambda model: model["id"] in app.state.config.MODEL_FILTER_LIST,
                         models["data"],
                     )
                 )
                 return models
         return models
     else:
-        url = app.state.OPENAI_API_BASE_URLS[url_idx]
+        url = app.state.config.OPENAI_API_BASE_URLS[url_idx]
 
         r = None
 
@@ -299,8 +335,8 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
     except json.JSONDecodeError as e:
         log.error("Error loading request body into a dictionary:", e)
 
-    url = app.state.OPENAI_API_BASE_URLS[idx]
-    key = app.state.OPENAI_API_KEYS[idx]
+    url = app.state.config.OPENAI_API_BASE_URLS[idx]
+    key = app.state.config.OPENAI_API_KEYS[idx]
 
     target_url = f"{url}/{path}"
 
@@ -341,7 +377,7 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
             try:
                 res = r.json()
                 if "error" in res:
-                    error_detail = f"External: {res['error']}"
+                    error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
             except:
                 error_detail = f"External: {e}"
 

backend/apps/rag/main.py

@@ -13,7 +13,6 @@ import os, shutil, logging, re
 from pathlib import Path
 from typing import List
 
-from chromadb.utils import embedding_functions
 from chromadb.utils.batch_utils import create_batches
 
 from langchain_community.document_loaders import (
@@ -29,15 +28,22 @@ from langchain_community.document_loaders import (
     UnstructuredXMLLoader,
     UnstructuredRSTLoader,
     UnstructuredExcelLoader,
+    YoutubeLoader,
 )
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 
+import validators
+import urllib.parse
+import socket
+
+
 from pydantic import BaseModel
 from typing import Optional
 import mimetypes
 import uuid
 import json
 
+import sentence_transformers
 
 from apps.web.models.documents import (
     Documents,
@@ -45,7 +51,14 @@ from apps.web.models.documents import (
     DocumentResponse,
 )
 
-from apps.rag.utils import query_doc, query_collection, get_embedding_model_path
+from apps.rag.utils import (
+    get_model_path,
+    get_embedding_function,
+    query_doc,
+    query_doc_with_hybrid_search,
+    query_collection,
+    query_collection_with_hybrid_search,
+)
 
 from utils.misc import (
     calculate_sha256,
@@ -54,17 +67,34 @@ from utils.misc import (
     extract_folders_after_data_docs,
 )
 from utils.utils import get_current_user, get_admin_user
+
 from config import (
+    ENV,
     SRC_LOG_LEVELS,
     UPLOAD_DIR,
     DOCS_DIR,
+    RAG_TOP_K,
+    RAG_RELEVANCE_THRESHOLD,
+    RAG_EMBEDDING_ENGINE,
     RAG_EMBEDDING_MODEL,
     RAG_EMBEDDING_MODEL_AUTO_UPDATE,
+    RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
+    ENABLE_RAG_HYBRID_SEARCH,
+    ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+    RAG_RERANKING_MODEL,
+    PDF_EXTRACT_IMAGES,
+    RAG_RERANKING_MODEL_AUTO_UPDATE,
+    RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
+    RAG_OPENAI_API_BASE_URL,
+    RAG_OPENAI_API_KEY,
     DEVICE_TYPE,
     CHROMA_CLIENT,
     CHUNK_SIZE,
     CHUNK_OVERLAP,
     RAG_TEMPLATE,
+    ENABLE_RAG_LOCAL_WEB_FETCH,
+    YOUTUBE_LOADER_LANGUAGE,
+    AppConfig,
 )
 
 from constants import ERROR_MESSAGES
@@ -74,29 +104,85 @@ log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 app = FastAPI()
 
-app.state.PDF_EXTRACT_IMAGES = False
-app.state.CHUNK_SIZE = CHUNK_SIZE
-app.state.CHUNK_OVERLAP = CHUNK_OVERLAP
-app.state.RAG_TEMPLATE = RAG_TEMPLATE
+app.state.config = AppConfig()
+
+app.state.config.TOP_K = RAG_TOP_K
+app.state.config.RELEVANCE_THRESHOLD = RAG_RELEVANCE_THRESHOLD
+
+app.state.config.ENABLE_RAG_HYBRID_SEARCH = ENABLE_RAG_HYBRID_SEARCH
+app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = (
+    ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION
+)
+
+app.state.config.CHUNK_SIZE = CHUNK_SIZE
+app.state.config.CHUNK_OVERLAP = CHUNK_OVERLAP
+
+app.state.config.RAG_EMBEDDING_ENGINE = RAG_EMBEDDING_ENGINE
+app.state.config.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL
+app.state.config.RAG_RERANKING_MODEL = RAG_RERANKING_MODEL
+app.state.config.RAG_TEMPLATE = RAG_TEMPLATE
 
 
-app.state.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL
+app.state.config.OPENAI_API_BASE_URL = RAG_OPENAI_API_BASE_URL
+app.state.config.OPENAI_API_KEY = RAG_OPENAI_API_KEY
+
+app.state.config.PDF_EXTRACT_IMAGES = PDF_EXTRACT_IMAGES
 
 
-app.state.TOP_K = 4
+app.state.config.YOUTUBE_LOADER_LANGUAGE = YOUTUBE_LOADER_LANGUAGE
+app.state.YOUTUBE_LOADER_TRANSLATION = None
 
-app.state.sentence_transformer_ef = (
-    embedding_functions.SentenceTransformerEmbeddingFunction(
-        model_name=get_embedding_model_path(
-            app.state.RAG_EMBEDDING_MODEL, RAG_EMBEDDING_MODEL_AUTO_UPDATE
-        ),
-        device=DEVICE_TYPE,
-    )
+
+def update_embedding_model(
+    embedding_model: str,
+    update_model: bool = False,
+):
+    if embedding_model and app.state.config.RAG_EMBEDDING_ENGINE == "":
+        app.state.sentence_transformer_ef = sentence_transformers.SentenceTransformer(
+            get_model_path(embedding_model, update_model),
+            device=DEVICE_TYPE,
+            trust_remote_code=RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
+        )
+    else:
+        app.state.sentence_transformer_ef = None
+
+
+def update_reranking_model(
+    reranking_model: str,
+    update_model: bool = False,
+):
+    if reranking_model:
+        app.state.sentence_transformer_rf = sentence_transformers.CrossEncoder(
+            get_model_path(reranking_model, update_model),
+            device=DEVICE_TYPE,
+            trust_remote_code=RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
+        )
+    else:
+        app.state.sentence_transformer_rf = None
+
+
+update_embedding_model(
+    app.state.config.RAG_EMBEDDING_MODEL,
+    RAG_EMBEDDING_MODEL_AUTO_UPDATE,
+)
+
+update_reranking_model(
+    app.state.config.RAG_RERANKING_MODEL,
+    RAG_RERANKING_MODEL_AUTO_UPDATE,
 )
 
 
+app.state.EMBEDDING_FUNCTION = get_embedding_function(
+    app.state.config.RAG_EMBEDDING_ENGINE,
+    app.state.config.RAG_EMBEDDING_MODEL,
+    app.state.sentence_transformer_ef,
+    app.state.config.OPENAI_API_KEY,
+    app.state.config.OPENAI_API_BASE_URL,
+)
+
 origins = ["*"]
 
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=origins,
@@ -110,7 +196,7 @@ class CollectionNameForm(BaseModel):
     collection_name: Optional[str] = "test"
 
 
-class StoreWebForm(CollectionNameForm):
+class UrlForm(CollectionNameForm):
     url: str
 
 
@@ -118,50 +204,82 @@ class StoreWebForm(CollectionNameForm):
 async def get_status():
     return {
         "status": True,
-        "chunk_size": app.state.CHUNK_SIZE,
-        "chunk_overlap": app.state.CHUNK_OVERLAP,
-        "template": app.state.RAG_TEMPLATE,
-        "embedding_model": app.state.RAG_EMBEDDING_MODEL,
+        "chunk_size": app.state.config.CHUNK_SIZE,
+        "chunk_overlap": app.state.config.CHUNK_OVERLAP,
+        "template": app.state.config.RAG_TEMPLATE,
+        "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE,
+        "embedding_model": app.state.config.RAG_EMBEDDING_MODEL,
+        "reranking_model": app.state.config.RAG_RERANKING_MODEL,
     }
 
 
-@app.get("/embedding/model")
-async def get_embedding_model(user=Depends(get_admin_user)):
+@app.get("/embedding")
+async def get_embedding_config(user=Depends(get_admin_user)):
     return {
         "status": True,
-        "embedding_model": app.state.RAG_EMBEDDING_MODEL,
+        "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE,
+        "embedding_model": app.state.config.RAG_EMBEDDING_MODEL,
+        "openai_config": {
+            "url": app.state.config.OPENAI_API_BASE_URL,
+            "key": app.state.config.OPENAI_API_KEY,
+        },
     }
 
 
+@app.get("/reranking")
+async def get_reraanking_config(user=Depends(get_admin_user)):
+    return {
+        "status": True,
+        "reranking_model": app.state.config.RAG_RERANKING_MODEL,
+    }
+
+
+class OpenAIConfigForm(BaseModel):
+    url: str
+    key: str
+
+
 class EmbeddingModelUpdateForm(BaseModel):
+    openai_config: Optional[OpenAIConfigForm] = None
+    embedding_engine: str
     embedding_model: str
 
 
-@app.post("/embedding/model/update")
-async def update_embedding_model(
+@app.post("/embedding/update")
+async def update_embedding_config(
     form_data: EmbeddingModelUpdateForm, user=Depends(get_admin_user)
 ):
-
     log.info(
-        f"Updating embedding model: {app.state.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}"
+        f"Updating embedding model: {app.state.config.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}"
     )
-
     try:
-        sentence_transformer_ef = (
-            embedding_functions.SentenceTransformerEmbeddingFunction(
-                model_name=get_embedding_model_path(form_data.embedding_model, True),
-                device=DEVICE_TYPE,
-            )
-        )
+        app.state.config.RAG_EMBEDDING_ENGINE = form_data.embedding_engine
+        app.state.config.RAG_EMBEDDING_MODEL = form_data.embedding_model
 
-        app.state.RAG_EMBEDDING_MODEL = form_data.embedding_model
-        app.state.sentence_transformer_ef = sentence_transformer_ef
+        if app.state.config.RAG_EMBEDDING_ENGINE in ["ollama", "openai"]:
+            if form_data.openai_config != None:
+                app.state.config.OPENAI_API_BASE_URL = form_data.openai_config.url
+                app.state.config.OPENAI_API_KEY = form_data.openai_config.key
+
+        update_embedding_model(app.state.config.RAG_EMBEDDING_MODEL)
+
+        app.state.EMBEDDING_FUNCTION = get_embedding_function(
+            app.state.config.RAG_EMBEDDING_ENGINE,
+            app.state.config.RAG_EMBEDDING_MODEL,
+            app.state.sentence_transformer_ef,
+            app.state.config.OPENAI_API_KEY,
+            app.state.config.OPENAI_API_BASE_URL,
+        )
 
         return {
             "status": True,
-            "embedding_model": app.state.RAG_EMBEDDING_MODEL,
+            "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE,
+            "embedding_model": app.state.config.RAG_EMBEDDING_MODEL,
+            "openai_config": {
+                "url": app.state.config.OPENAI_API_BASE_URL,
+                "key": app.state.config.OPENAI_API_KEY,
+            },
         }
-
     except Exception as e:
         log.exception(f"Problem updating embedding model: {e}")
         raise HTTPException(
@@ -170,14 +288,47 @@ async def update_embedding_model(
         )
 
 
+class RerankingModelUpdateForm(BaseModel):
+    reranking_model: str
+
+
+@app.post("/reranking/update")
+async def update_reranking_config(
+    form_data: RerankingModelUpdateForm, user=Depends(get_admin_user)
+):
+    log.info(
+        f"Updating reranking model: {app.state.config.RAG_RERANKING_MODEL} to {form_data.reranking_model}"
+    )
+    try:
+        app.state.config.RAG_RERANKING_MODEL = form_data.reranking_model
+
+        update_reranking_model(app.state.config.RAG_RERANKING_MODEL), True
+
+        return {
+            "status": True,
+            "reranking_model": app.state.config.RAG_RERANKING_MODEL,
+        }
+    except Exception as e:
+        log.exception(f"Problem updating reranking model: {e}")
+        raise HTTPException(
+            status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
 @app.get("/config")
 async def get_rag_config(user=Depends(get_admin_user)):
     return {
         "status": True,
-        "pdf_extract_images": app.state.PDF_EXTRACT_IMAGES,
+        "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES,
         "chunk": {
-            "chunk_size": app.state.CHUNK_SIZE,
-            "chunk_overlap": app.state.CHUNK_OVERLAP,
+            "chunk_size": app.state.config.CHUNK_SIZE,
+            "chunk_overlap": app.state.config.CHUNK_OVERLAP,
+        },
+        "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+        "youtube": {
+            "language": app.state.config.YOUTUBE_LOADER_LANGUAGE,
+            "translation": app.state.YOUTUBE_LOADER_TRANSLATION,
         },
     }
 
@@ -187,23 +338,67 @@ class ChunkParamUpdateForm(BaseModel):
     chunk_overlap: int
 
 
+class YoutubeLoaderConfig(BaseModel):
+    language: List[str]
+    translation: Optional[str] = None
+
+
 class ConfigUpdateForm(BaseModel):
-    pdf_extract_images: bool
-    chunk: ChunkParamUpdateForm
+    pdf_extract_images: Optional[bool] = None
+    chunk: Optional[ChunkParamUpdateForm] = None
+    web_loader_ssl_verification: Optional[bool] = None
+    youtube: Optional[YoutubeLoaderConfig] = None
 
 
 @app.post("/config/update")
 async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_user)):
-    app.state.PDF_EXTRACT_IMAGES = form_data.pdf_extract_images
-    app.state.CHUNK_SIZE = form_data.chunk.chunk_size
-    app.state.CHUNK_OVERLAP = form_data.chunk.chunk_overlap
+    app.state.config.PDF_EXTRACT_IMAGES = (
+        form_data.pdf_extract_images
+        if form_data.pdf_extract_images is not None
+        else app.state.config.PDF_EXTRACT_IMAGES
+    )
+
+    app.state.config.CHUNK_SIZE = (
+        form_data.chunk.chunk_size
+        if form_data.chunk is not None
+        else app.state.config.CHUNK_SIZE
+    )
+
+    app.state.config.CHUNK_OVERLAP = (
+        form_data.chunk.chunk_overlap
+        if form_data.chunk is not None
+        else app.state.config.CHUNK_OVERLAP
+    )
+
+    app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = (
+        form_data.web_loader_ssl_verification
+        if form_data.web_loader_ssl_verification != None
+        else app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION
+    )
+
+    app.state.config.YOUTUBE_LOADER_LANGUAGE = (
+        form_data.youtube.language
+        if form_data.youtube is not None
+        else app.state.config.YOUTUBE_LOADER_LANGUAGE
+    )
+
+    app.state.YOUTUBE_LOADER_TRANSLATION = (
+        form_data.youtube.translation
+        if form_data.youtube is not None
+        else app.state.YOUTUBE_LOADER_TRANSLATION
+    )
 
     return {
         "status": True,
-        "pdf_extract_images": app.state.PDF_EXTRACT_IMAGES,
+        "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES,
         "chunk": {
-            "chunk_size": app.state.CHUNK_SIZE,
-            "chunk_overlap": app.state.CHUNK_OVERLAP,
+            "chunk_size": app.state.config.CHUNK_SIZE,
+            "chunk_overlap": app.state.config.CHUNK_OVERLAP,
+        },
+        "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+        "youtube": {
+            "language": app.state.config.YOUTUBE_LOADER_LANGUAGE,
+            "translation": app.state.YOUTUBE_LOADER_TRANSLATION,
         },
     }
 
@@ -212,7 +407,7 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
 async def get_rag_template(user=Depends(get_current_user)):
     return {
         "status": True,
-        "template": app.state.RAG_TEMPLATE,
+        "template": app.state.config.RAG_TEMPLATE,
     }
 
 
@@ -220,29 +415,47 @@ async def get_rag_template(user=Depends(get_current_user)):
 async def get_query_settings(user=Depends(get_admin_user)):
     return {
         "status": True,
-        "template": app.state.RAG_TEMPLATE,
-        "k": app.state.TOP_K,
+        "template": app.state.config.RAG_TEMPLATE,
+        "k": app.state.config.TOP_K,
+        "r": app.state.config.RELEVANCE_THRESHOLD,
+        "hybrid": app.state.config.ENABLE_RAG_HYBRID_SEARCH,
     }
 
 
 class QuerySettingsForm(BaseModel):
     k: Optional[int] = None
+    r: Optional[float] = None
     template: Optional[str] = None
+    hybrid: Optional[bool] = None
 
 
 @app.post("/query/settings/update")
 async def update_query_settings(
     form_data: QuerySettingsForm, user=Depends(get_admin_user)
 ):
-    app.state.RAG_TEMPLATE = form_data.template if form_data.template else RAG_TEMPLATE
-    app.state.TOP_K = form_data.k if form_data.k else 4
-    return {"status": True, "template": app.state.RAG_TEMPLATE}
+    app.state.config.RAG_TEMPLATE = (
+        form_data.template if form_data.template else RAG_TEMPLATE
+    )
+    app.state.config.TOP_K = form_data.k if form_data.k else 4
+    app.state.config.RELEVANCE_THRESHOLD = form_data.r if form_data.r else 0.0
+    app.state.config.ENABLE_RAG_HYBRID_SEARCH = (
+        form_data.hybrid if form_data.hybrid else False
+    )
+    return {
+        "status": True,
+        "template": app.state.config.RAG_TEMPLATE,
+        "k": app.state.config.TOP_K,
+        "r": app.state.config.RELEVANCE_THRESHOLD,
+        "hybrid": app.state.config.ENABLE_RAG_HYBRID_SEARCH,
+    }
 
 
 class QueryDocForm(BaseModel):
     collection_name: str
     query: str
     k: Optional[int] = None
+    r: Optional[float] = None
+    hybrid: Optional[bool] = None
 
 
 @app.post("/query/doc")
@@ -250,14 +463,25 @@ def query_doc_handler(
     form_data: QueryDocForm,
     user=Depends(get_current_user),
 ):
-
     try:
-        return query_doc(
-            collection_name=form_data.collection_name,
-            query=form_data.query,
-            k=form_data.k if form_data.k else app.state.TOP_K,
-            embedding_function=app.state.sentence_transformer_ef,
-        )
+        if app.state.config.ENABLE_RAG_HYBRID_SEARCH:
+            return query_doc_with_hybrid_search(
+                collection_name=form_data.collection_name,
+                query=form_data.query,
+                embedding_function=app.state.EMBEDDING_FUNCTION,
+                k=form_data.k if form_data.k else app.state.config.TOP_K,
+                reranking_function=app.state.sentence_transformer_rf,
+                r=(
+                    form_data.r if form_data.r else app.state.config.RELEVANCE_THRESHOLD
+                ),
+            )
+        else:
+            return query_doc(
+                collection_name=form_data.collection_name,
+                query=form_data.query,
+                embedding_function=app.state.EMBEDDING_FUNCTION,
+                k=form_data.k if form_data.k else app.state.config.TOP_K,
+            )
     except Exception as e:
         log.exception(e)
         raise HTTPException(
@@ -270,6 +494,8 @@ class QueryCollectionsForm(BaseModel):
     collection_names: List[str]
     query: str
     k: Optional[int] = None
+    r: Optional[float] = None
+    hybrid: Optional[bool] = None
 
 
 @app.post("/query/collection")
@@ -277,19 +503,43 @@ def query_collection_handler(
     form_data: QueryCollectionsForm,
     user=Depends(get_current_user),
 ):
-    return query_collection(
-        collection_names=form_data.collection_names,
-        query=form_data.query,
-        k=form_data.k if form_data.k else app.state.TOP_K,
-        embedding_function=app.state.sentence_transformer_ef,
-    )
-
-
-@app.post("/web")
-def store_web(form_data: StoreWebForm, user=Depends(get_current_user)):
-    # "https://www.gutenberg.org/files/1727/1727-h/1727-h.htm"
     try:
-        loader = WebBaseLoader(form_data.url)
+        if app.state.config.ENABLE_RAG_HYBRID_SEARCH:
+            return query_collection_with_hybrid_search(
+                collection_names=form_data.collection_names,
+                query=form_data.query,
+                embedding_function=app.state.EMBEDDING_FUNCTION,
+                k=form_data.k if form_data.k else app.state.config.TOP_K,
+                reranking_function=app.state.sentence_transformer_rf,
+                r=(
+                    form_data.r if form_data.r else app.state.config.RELEVANCE_THRESHOLD
+                ),
+            )
+        else:
+            return query_collection(
+                collection_names=form_data.collection_names,
+                query=form_data.query,
+                embedding_function=app.state.EMBEDDING_FUNCTION,
+                k=form_data.k if form_data.k else app.state.config.TOP_K,
+            )
+
+    except Exception as e:
+        log.exception(e)
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+@app.post("/youtube")
+def store_youtube_video(form_data: UrlForm, user=Depends(get_current_user)):
+    try:
+        loader = YoutubeLoader.from_youtube_url(
+            form_data.url,
+            add_video_info=True,
+            language=app.state.config.YOUTUBE_LOADER_LANGUAGE,
+            translation=app.state.YOUTUBE_LOADER_TRANSLATION,
+        )
         data = loader.load()
 
         collection_name = form_data.collection_name
@@ -310,16 +560,77 @@ def store_web(form_data: StoreWebForm, user=Depends(get_current_user)):
         )
 
 
+@app.post("/web")
+def store_web(form_data: UrlForm, user=Depends(get_current_user)):
+    # "https://www.gutenberg.org/files/1727/1727-h/1727-h.htm"
+    try:
+        loader = get_web_loader(
+            form_data.url,
+            verify_ssl=app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+        )
+        data = loader.load()
+
+        collection_name = form_data.collection_name
+        if collection_name == "":
+            collection_name = calculate_sha256_string(form_data.url)[:63]
+
+        store_data_in_vector_db(data, collection_name, overwrite=True)
+        return {
+            "status": True,
+            "collection_name": collection_name,
+            "filename": form_data.url,
+        }
+    except Exception as e:
+        log.exception(e)
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+def get_web_loader(url: str, verify_ssl: bool = True):
+    # Check if the URL is valid
+    if isinstance(validators.url(url), validators.ValidationError):
+        raise ValueError(ERROR_MESSAGES.INVALID_URL)
+    if not ENABLE_RAG_LOCAL_WEB_FETCH:
+        # Local web fetch is disabled, filter out any URLs that resolve to private IP addresses
+        parsed_url = urllib.parse.urlparse(url)
+        # Get IPv4 and IPv6 addresses
+        ipv4_addresses, ipv6_addresses = resolve_hostname(parsed_url.hostname)
+        # Check if any of the resolved addresses are private
+        # This is technically still vulnerable to DNS rebinding attacks, as we don't control WebBaseLoader
+        for ip in ipv4_addresses:
+            if validators.ipv4(ip, private=True):
+                raise ValueError(ERROR_MESSAGES.INVALID_URL)
+        for ip in ipv6_addresses:
+            if validators.ipv6(ip, private=True):
+                raise ValueError(ERROR_MESSAGES.INVALID_URL)
+    return WebBaseLoader(url, verify_ssl=verify_ssl)
+
+
+def resolve_hostname(hostname):
+    # Get address information
+    addr_info = socket.getaddrinfo(hostname, None)
+
+    # Extract IP addresses from address information
+    ipv4_addresses = [info[4][0] for info in addr_info if info[0] == socket.AF_INET]
+    ipv6_addresses = [info[4][0] for info in addr_info if info[0] == socket.AF_INET6]
+
+    return ipv4_addresses, ipv6_addresses
+
+
 def store_data_in_vector_db(data, collection_name, overwrite: bool = False) -> bool:
 
     text_splitter = RecursiveCharacterTextSplitter(
-        chunk_size=app.state.CHUNK_SIZE,
-        chunk_overlap=app.state.CHUNK_OVERLAP,
+        chunk_size=app.state.config.CHUNK_SIZE,
+        chunk_overlap=app.state.config.CHUNK_OVERLAP,
         add_start_index=True,
     )
+
     docs = text_splitter.split_documents(data)
 
     if len(docs) > 0:
+        log.info(f"store_data_in_vector_db {docs}")
         return store_docs_in_vector_db(docs, collection_name, overwrite), None
     else:
         raise ValueError(ERROR_MESSAGES.EMPTY_CONTENT)
@@ -329,8 +640,8 @@ def store_text_in_vector_db(
     text, metadata, collection_name, overwrite: bool = False
 ) -> bool:
     text_splitter = RecursiveCharacterTextSplitter(
-        chunk_size=app.state.CHUNK_SIZE,
-        chunk_overlap=app.state.CHUNK_OVERLAP,
+        chunk_size=app.state.config.CHUNK_SIZE,
+        chunk_overlap=app.state.config.CHUNK_OVERLAP,
         add_start_index=True,
     )
     docs = text_splitter.create_documents([text], metadatas=[metadata])
@@ -338,6 +649,7 @@ def store_text_in_vector_db(
 
 
 def store_docs_in_vector_db(docs, collection_name, overwrite: bool = False) -> bool:
+    log.info(f"store_docs_in_vector_db {docs} {collection_name}")
 
     texts = [doc.page_content for doc in docs]
     metadatas = [doc.metadata for doc in docs]
@@ -349,15 +661,24 @@ def store_docs_in_vector_db(docs, collection_name, overwrite: bool = False) -> b
                     log.info(f"deleting existing collection {collection_name}")
                     CHROMA_CLIENT.delete_collection(name=collection_name)
 
-        collection = CHROMA_CLIENT.create_collection(
-            name=collection_name,
-            embedding_function=app.state.sentence_transformer_ef,
+        collection = CHROMA_CLIENT.create_collection(name=collection_name)
+
+        embedding_func = get_embedding_function(
+            app.state.config.RAG_EMBEDDING_ENGINE,
+            app.state.config.RAG_EMBEDDING_MODEL,
+            app.state.sentence_transformer_ef,
+            app.state.config.OPENAI_API_KEY,
+            app.state.config.OPENAI_API_BASE_URL,
         )
 
+        embedding_texts = list(map(lambda x: x.replace("\n", " "), texts))
+        embeddings = embedding_func(embedding_texts)
+
         for batch in create_batches(
             api=CHROMA_CLIENT,
-            ids=[str(uuid.uuid1()) for _ in texts],
+            ids=[str(uuid.uuid4()) for _ in texts],
             metadatas=metadatas,
+            embeddings=embeddings,
             documents=texts,
         ):
             collection.add(*batch)
@@ -421,7 +742,9 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
     ]
 
     if file_ext == "pdf":
-        loader = PyPDFLoader(file_path, extract_images=app.state.PDF_EXTRACT_IMAGES)
+        loader = PyPDFLoader(
+            file_path, extract_images=app.state.config.PDF_EXTRACT_IMAGES
+        )
     elif file_ext == "csv":
         loader = CSVLoader(file_path)
     elif file_ext == "rst":
@@ -629,3 +952,14 @@ def reset(user=Depends(get_admin_user)) -> bool:
         log.exception(e)
 
     return True
+
+
+if ENV == "dev":
+
+    @app.get("/ef")
+    async def get_embeddings():
+        return {"result": app.state.EMBEDDING_FUNCTION("hello world")}
+
+    @app.get("/ef/{text}")
+    async def get_embeddings_text(text: str):
+        return {"result": app.state.EMBEDDING_FUNCTION(text)}

backend/apps/rag/utils.py

@@ -1,99 +1,189 @@
 import os
-import re
 import logging
+import requests
+
 from typing import List
+
+from apps.ollama.main import (
+    generate_ollama_embeddings,
+    GenerateEmbeddingsForm,
+)
+
 from huggingface_hub import snapshot_download
 
+from langchain_core.documents import Document
+from langchain_community.retrievers import BM25Retriever
+from langchain.retrievers import (
+    ContextualCompressionRetriever,
+    EnsembleRetriever,
+)
+
+from typing import Optional
 from config import SRC_LOG_LEVELS, CHROMA_CLIENT
 
+
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
-def query_doc(collection_name: str, query: str, k: int, embedding_function):
+def query_doc(
+    collection_name: str,
+    query: str,
+    embedding_function,
+    k: int,
+):
     try:
-        # if you use docker use the model from the environment variable
-        collection = CHROMA_CLIENT.get_collection(
-            name=collection_name,
-            embedding_function=embedding_function,
-        )
+        collection = CHROMA_CLIENT.get_collection(name=collection_name)
+        query_embeddings = embedding_function(query)
+
         result = collection.query(
-            query_texts=[query],
+            query_embeddings=[query_embeddings],
             n_results=k,
         )
+
+        log.info(f"query_doc:result {result}")
         return result
     except Exception as e:
         raise e
 
 
-def merge_and_sort_query_results(query_results, k):
+def query_doc_with_hybrid_search(
+    collection_name: str,
+    query: str,
+    embedding_function,
+    k: int,
+    reranking_function,
+    r: float,
+):
+    try:
+        collection = CHROMA_CLIENT.get_collection(name=collection_name)
+        documents = collection.get()  # get all documents
+
+        bm25_retriever = BM25Retriever.from_texts(
+            texts=documents.get("documents"),
+            metadatas=documents.get("metadatas"),
+        )
+        bm25_retriever.k = k
+
+        chroma_retriever = ChromaRetriever(
+            collection=collection,
+            embedding_function=embedding_function,
+            top_n=k,
+        )
+
+        ensemble_retriever = EnsembleRetriever(
+            retrievers=[bm25_retriever, chroma_retriever], weights=[0.5, 0.5]
+        )
+
+        compressor = RerankCompressor(
+            embedding_function=embedding_function,
+            top_n=k,
+            reranking_function=reranking_function,
+            r_score=r,
+        )
+
+        compression_retriever = ContextualCompressionRetriever(
+            base_compressor=compressor, base_retriever=ensemble_retriever
+        )
+
+        result = compression_retriever.invoke(query)
+        result = {
+            "distances": [[d.metadata.get("score") for d in result]],
+            "documents": [[d.page_content for d in result]],
+            "metadatas": [[d.metadata for d in result]],
+        }
+
+        log.info(f"query_doc_with_hybrid_search:result {result}")
+        return result
+    except Exception as e:
+        raise e
+
+
+def merge_and_sort_query_results(query_results, k, reverse=False):
     # Initialize lists to store combined data
-    combined_ids = []
     combined_distances = []
-    combined_metadatas = []
     combined_documents = []
+    combined_metadatas = []
 
-    # Combine data from each dictionary
     for data in query_results:
-        combined_ids.extend(data["ids"][0])
         combined_distances.extend(data["distances"][0])
-        combined_metadatas.extend(data["metadatas"][0])
         combined_documents.extend(data["documents"][0])
+        combined_metadatas.extend(data["metadatas"][0])
 
-    # Create a list of tuples (distance, id, metadata, document)
-    combined = list(
-        zip(combined_distances, combined_ids, combined_metadatas, combined_documents)
-    )
+    # Create a list of tuples (distance, document, metadata)
+    combined = list(zip(combined_distances, combined_documents, combined_metadatas))
 
     # Sort the list based on distances
-    combined.sort(key=lambda x: x[0])
+    combined.sort(key=lambda x: x[0], reverse=reverse)
 
-    # Unzip the sorted list
-    sorted_distances, sorted_ids, sorted_metadatas, sorted_documents = zip(*combined)
+    # We don't have anything :-(
+    if not combined:
+        sorted_distances = []
+        sorted_documents = []
+        sorted_metadatas = []
+    else:
+        # Unzip the sorted list
+        sorted_distances, sorted_documents, sorted_metadatas = zip(*combined)
 
-    # Slicing the lists to include only k elements
-    sorted_distances = list(sorted_distances)[:k]
-    sorted_ids = list(sorted_ids)[:k]
-    sorted_metadatas = list(sorted_metadatas)[:k]
-    sorted_documents = list(sorted_documents)[:k]
+        # Slicing the lists to include only k elements
+        sorted_distances = list(sorted_distances)[:k]
+        sorted_documents = list(sorted_documents)[:k]
+        sorted_metadatas = list(sorted_metadatas)[:k]
 
     # Create the output dictionary
-    merged_query_results = {
-        "ids": [sorted_ids],
+    result = {
         "distances": [sorted_distances],
-        "metadatas": [sorted_metadatas],
         "documents": [sorted_documents],
-        "embeddings": None,
-        "uris": None,
-        "data": None,
+        "metadatas": [sorted_metadatas],
     }
 
-    return merged_query_results
+    return result
 
 
 def query_collection(
-    collection_names: List[str], query: str, k: int, embedding_function
+    collection_names: List[str],
+    query: str,
+    embedding_function,
+    k: int,
 ):
-
     results = []
-
     for collection_name in collection_names:
         try:
-            # if you use docker use the model from the environment variable
-            collection = CHROMA_CLIENT.get_collection(
-                name=collection_name,
+            result = query_doc(
+                collection_name=collection_name,
+                query=query,
+                k=k,
                 embedding_function=embedding_function,
             )
-
-            result = collection.query(
-                query_texts=[query],
-                n_results=k,
-            )
             results.append(result)
         except:
             pass
+    return merge_and_sort_query_results(results, k=k)
 
-    return merge_and_sort_query_results(results, k)
+
+def query_collection_with_hybrid_search(
+    collection_names: List[str],
+    query: str,
+    embedding_function,
+    k: int,
+    reranking_function,
+    r: float,
+):
+    results = []
+    for collection_name in collection_names:
+        try:
+            result = query_doc_with_hybrid_search(
+                collection_name=collection_name,
+                query=query,
+                embedding_function=embedding_function,
+                k=k,
+                reranking_function=reranking_function,
+                r=r,
+            )
+            results.append(result)
+        except:
+            pass
+    return merge_and_sort_query_results(results, k=k, reverse=True)
 
 
 def rag_template(template: str, context: str, query: str):
@@ -102,8 +192,53 @@ def rag_template(template: str, context: str, query: str):
     return template
 
 
-def rag_messages(docs, messages, template, k, embedding_function):
-    log.debug(f"docs: {docs}")
+def get_embedding_function(
+    embedding_engine,
+    embedding_model,
+    embedding_function,
+    openai_key,
+    openai_url,
+):
+    if embedding_engine == "":
+        return lambda query: embedding_function.encode(query).tolist()
+    elif embedding_engine in ["ollama", "openai"]:
+        if embedding_engine == "ollama":
+            func = lambda query: generate_ollama_embeddings(
+                GenerateEmbeddingsForm(
+                    **{
+                        "model": embedding_model,
+                        "prompt": query,
+                    }
+                )
+            )
+        elif embedding_engine == "openai":
+            func = lambda query: generate_openai_embeddings(
+                model=embedding_model,
+                text=query,
+                key=openai_key,
+                url=openai_url,
+            )
+
+        def generate_multiple(query, f):
+            if isinstance(query, list):
+                return [f(q) for q in query]
+            else:
+                return f(query)
+
+        return lambda query: generate_multiple(query, func)
+
+
+def rag_messages(
+    docs,
+    messages,
+    template,
+    embedding_function,
+    k,
+    reranking_function,
+    r,
+    hybrid_search,
+):
+    log.debug(f"docs: {docs} {messages} {embedding_function} {reranking_function}")
 
     last_user_message_idx = None
     for i in range(len(messages) - 1, -1, -1):
@@ -130,40 +265,74 @@ def rag_messages(docs, messages, template, k, embedding_function):
         content_type = None
         query = ""
 
+    extracted_collections = []
     relevant_contexts = []
 
     for doc in docs:
         context = None
 
+        collection_names = (
+            doc["collection_names"]
+            if doc["type"] == "collection"
+            else [doc["collection_name"]]
+        )
+
+        collection_names = set(collection_names).difference(extracted_collections)
+        if not collection_names:
+            log.debug(f"skipping {doc} as it has already been extracted")
+            continue
+
         try:
-            if doc["type"] == "collection":
-                context = query_collection(
-                    collection_names=doc["collection_names"],
-                    query=query,
-                    k=k,
-                    embedding_function=embedding_function,
-                )
-            elif doc["type"] == "text":
+            if doc["type"] == "text":
                 context = doc["content"]
             else:
-                context = query_doc(
-                    collection_name=doc["collection_name"],
-                    query=query,
-                    k=k,
-                    embedding_function=embedding_function,
-                )
+                if hybrid_search:
+                    context = query_collection_with_hybrid_search(
+                        collection_names=collection_names,
+                        query=query,
+                        embedding_function=embedding_function,
+                        k=k,
+                        reranking_function=reranking_function,
+                        r=r,
+                    )
+                else:
+                    context = query_collection(
+                        collection_names=collection_names,
+                        query=query,
+                        embedding_function=embedding_function,
+                        k=k,
+                    )
         except Exception as e:
             log.exception(e)
             context = None
 
-        relevant_contexts.append(context)
+        if context:
+            relevant_contexts.append({**context, "source": doc})
 
-    log.debug(f"relevant_contexts: {relevant_contexts}")
+        extracted_collections.extend(collection_names)
 
     context_string = ""
+
+    citations = []
     for context in relevant_contexts:
-        if context:
-            context_string += " ".join(context["documents"][0]) + "\n"
+        try:
+            if "documents" in context:
+                context_string += "\n\n".join(
+                    [text for text in context["documents"][0] if text is not None]
+                )
+
+                if "metadatas" in context:
+                    citations.append(
+                        {
+                            "source": context["source"],
+                            "document": context["documents"][0],
+                            "metadata": context["metadatas"][0],
+                        }
+                    )
+        except Exception as e:
+            log.exception(e)
+
+    context_string = context_string.strip()
 
     ra_content = rag_template(
         template=template,
@@ -171,6 +340,8 @@ def rag_messages(docs, messages, template, k, embedding_function):
         query=query,
     )
 
+    log.debug(f"ra_content: {ra_content}")
+
     if content_type == "list":
         new_content = []
         for content_item in user_message["content"]:
@@ -189,44 +360,163 @@ def rag_messages(docs, messages, template, k, embedding_function):
 
     messages[last_user_message_idx] = new_user_message
 
-    return messages
+    return messages, citations
 
 
-def get_embedding_model_path(
-    embedding_model: str, update_embedding_model: bool = False
-):
+def get_model_path(model: str, update_model: bool = False):
     # Construct huggingface_hub kwargs with local_files_only to return the snapshot path
     cache_dir = os.getenv("SENTENCE_TRANSFORMERS_HOME")
 
-    local_files_only = not update_embedding_model
+    local_files_only = not update_model
 
     snapshot_kwargs = {
         "cache_dir": cache_dir,
         "local_files_only": local_files_only,
     }
 
-    log.debug(f"embedding_model: {embedding_model}")
+    log.debug(f"model: {model}")
     log.debug(f"snapshot_kwargs: {snapshot_kwargs}")
 
     # Inspiration from upstream sentence_transformers
     if (
-        os.path.exists(embedding_model)
-        or ("\\" in embedding_model or embedding_model.count("/") > 1)
+        os.path.exists(model)
+        or ("\\" in model or model.count("/") > 1)
         and local_files_only
     ):
         # If fully qualified path exists, return input, else set repo_id
-        return embedding_model
-    elif "/" not in embedding_model:
+        return model
+    elif "/" not in model:
         # Set valid repo_id for model short-name
-        embedding_model = "sentence-transformers" + "/" + embedding_model
+        model = "sentence-transformers" + "/" + model
 
-    snapshot_kwargs["repo_id"] = embedding_model
+    snapshot_kwargs["repo_id"] = model
 
     # Attempt to query the huggingface_hub library to determine the local path and/or to update
     try:
-        embedding_model_repo_path = snapshot_download(**snapshot_kwargs)
-        log.debug(f"embedding_model_repo_path: {embedding_model_repo_path}")
-        return embedding_model_repo_path
+        model_repo_path = snapshot_download(**snapshot_kwargs)
+        log.debug(f"model_repo_path: {model_repo_path}")
+        return model_repo_path
     except Exception as e:
-        log.exception(f"Cannot determine embedding model snapshot path: {e}")
-        return embedding_model
+        log.exception(f"Cannot determine model snapshot path: {e}")
+        return model
+
+
+def generate_openai_embeddings(
+    model: str, text: str, key: str, url: str = "https://api.openai.com/v1"
+):
+    try:
+        r = requests.post(
+            f"{url}/embeddings",
+            headers={
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {key}",
+            },
+            json={"input": text, "model": model},
+        )
+        r.raise_for_status()
+        data = r.json()
+        if "data" in data:
+            return data["data"][0]["embedding"]
+        else:
+            raise "Something went wrong :/"
+    except Exception as e:
+        print(e)
+        return None
+
+
+from typing import Any
+
+from langchain_core.retrievers import BaseRetriever
+from langchain_core.callbacks import CallbackManagerForRetrieverRun
+
+
+class ChromaRetriever(BaseRetriever):
+    collection: Any
+    embedding_function: Any
+    top_n: int
+
+    def _get_relevant_documents(
+        self,
+        query: str,
+        *,
+        run_manager: CallbackManagerForRetrieverRun,
+    ) -> List[Document]:
+        query_embeddings = self.embedding_function(query)
+
+        results = self.collection.query(
+            query_embeddings=[query_embeddings],
+            n_results=self.top_n,
+        )
+
+        ids = results["ids"][0]
+        metadatas = results["metadatas"][0]
+        documents = results["documents"][0]
+
+        results = []
+        for idx in range(len(ids)):
+            results.append(
+                Document(
+                    metadata=metadatas[idx],
+                    page_content=documents[idx],
+                )
+            )
+        return results
+
+
+import operator
+
+from typing import Optional, Sequence
+
+from langchain_core.documents import BaseDocumentCompressor, Document
+from langchain_core.callbacks import Callbacks
+from langchain_core.pydantic_v1 import Extra
+
+from sentence_transformers import util
+
+
+class RerankCompressor(BaseDocumentCompressor):
+    embedding_function: Any
+    top_n: int
+    reranking_function: Any
+    r_score: float
+
+    class Config:
+        extra = Extra.forbid
+        arbitrary_types_allowed = True
+
+    def compress_documents(
+        self,
+        documents: Sequence[Document],
+        query: str,
+        callbacks: Optional[Callbacks] = None,
+    ) -> Sequence[Document]:
+        reranking = self.reranking_function is not None
+
+        if reranking:
+            scores = self.reranking_function.predict(
+                [(query, doc.page_content) for doc in documents]
+            )
+        else:
+            query_embedding = self.embedding_function(query)
+            document_embedding = self.embedding_function(
+                [doc.page_content for doc in documents]
+            )
+            scores = util.cos_sim(query_embedding, document_embedding)[0]
+
+        docs_with_scores = list(zip(documents, scores.tolist()))
+        if self.r_score:
+            docs_with_scores = [
+                (d, s) for d, s in docs_with_scores if s >= self.r_score
+            ]
+
+        result = sorted(docs_with_scores, key=operator.itemgetter(1), reverse=True)
+        final_results = []
+        for doc, doc_score in result[: self.top_n]:
+            metadata = doc.metadata
+            metadata["score"] = doc_score
+            doc = Document(
+                page_content=doc.page_content,
+                metadata=metadata,
+            )
+            final_results.append(doc)
+        return final_results

backend/apps/web/internal/db.py

@@ -1,6 +1,7 @@
 from peewee import *
 from peewee_migrate import Router
-from config import SRC_LOG_LEVELS, DATA_DIR
+from playhouse.db_url import connect
+from config import SRC_LOG_LEVELS, DATA_DIR, DATABASE_URL
 import os
 import logging
 
@@ -11,12 +12,12 @@ log.setLevel(SRC_LOG_LEVELS["DB"])
 if os.path.exists(f"{DATA_DIR}/ollama.db"):
     # Rename the file
     os.rename(f"{DATA_DIR}/ollama.db", f"{DATA_DIR}/webui.db")
-    log.info("File renamed successfully.")
+    log.info("Database migrated from Ollama-WebUI successfully.")
 else:
     pass
 
-
-DB = SqliteDatabase(f"{DATA_DIR}/webui.db")
+DB = connect(DATABASE_URL)
+log.info(f"Connected to a {DB.__class__.__name__} database.")
 router = Router(DB, migrate_dir="apps/web/internal/migrations", logger=log)
 router.run()
 DB.connect(reuse_if_open=True)

backend/apps/web/internal/migrations/001_initial_schema.py

@@ -37,6 +37,18 @@ with suppress(ImportError):
 def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
     """Write your migrations here."""
 
+    # We perform different migrations for SQLite and other databases
+    # This is because SQLite is very loose with enforcing its schema, and trying to migrate other databases like SQLite
+    # will require per-database SQL queries.
+    # Instead, we assume that because external DB support was added at a later date, it is safe to assume a newer base
+    # schema instead of trying to migrate from an older schema.
+    if isinstance(database, pw.SqliteDatabase):
+        migrate_sqlite(migrator, database, fake=fake)
+    else:
+        migrate_external(migrator, database, fake=fake)
+
+
+def migrate_sqlite(migrator: Migrator, database: pw.Database, *, fake=False):
     @migrator.create_model
     class Auth(pw.Model):
         id = pw.CharField(max_length=255, unique=True)
@@ -53,7 +65,7 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         user_id = pw.CharField(max_length=255)
         title = pw.CharField()
         chat = pw.TextField()
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "chat"
@@ -64,7 +76,7 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         tag_name = pw.CharField(max_length=255)
         chat_id = pw.CharField(max_length=255)
         user_id = pw.CharField(max_length=255)
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "chatidtag"
@@ -78,7 +90,7 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         filename = pw.CharField()
         content = pw.TextField(null=True)
         user_id = pw.CharField(max_length=255)
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "document"
@@ -89,7 +101,7 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         tag_name = pw.CharField(max_length=255, unique=True)
         user_id = pw.CharField(max_length=255)
         modelfile = pw.TextField()
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "modelfile"
@@ -101,7 +113,7 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         user_id = pw.CharField(max_length=255)
         title = pw.CharField()
         content = pw.TextField()
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "prompt"
@@ -123,7 +135,100 @@ def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
         email = pw.CharField(max_length=255)
         role = pw.CharField(max_length=255)
         profile_image_url = pw.CharField(max_length=255)
-        timestamp = pw.DateField()
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "user"
+
+
+def migrate_external(migrator: Migrator, database: pw.Database, *, fake=False):
+    @migrator.create_model
+    class Auth(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        email = pw.CharField(max_length=255)
+        password = pw.TextField()
+        active = pw.BooleanField()
+
+        class Meta:
+            table_name = "auth"
+
+    @migrator.create_model
+    class Chat(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        user_id = pw.CharField(max_length=255)
+        title = pw.TextField()
+        chat = pw.TextField()
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "chat"
+
+    @migrator.create_model
+    class ChatIdTag(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        tag_name = pw.CharField(max_length=255)
+        chat_id = pw.CharField(max_length=255)
+        user_id = pw.CharField(max_length=255)
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "chatidtag"
+
+    @migrator.create_model
+    class Document(pw.Model):
+        id = pw.AutoField()
+        collection_name = pw.CharField(max_length=255, unique=True)
+        name = pw.CharField(max_length=255, unique=True)
+        title = pw.TextField()
+        filename = pw.TextField()
+        content = pw.TextField(null=True)
+        user_id = pw.CharField(max_length=255)
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "document"
+
+    @migrator.create_model
+    class Modelfile(pw.Model):
+        id = pw.AutoField()
+        tag_name = pw.CharField(max_length=255, unique=True)
+        user_id = pw.CharField(max_length=255)
+        modelfile = pw.TextField()
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "modelfile"
+
+    @migrator.create_model
+    class Prompt(pw.Model):
+        id = pw.AutoField()
+        command = pw.CharField(max_length=255, unique=True)
+        user_id = pw.CharField(max_length=255)
+        title = pw.TextField()
+        content = pw.TextField()
+        timestamp = pw.BigIntegerField()
+
+        class Meta:
+            table_name = "prompt"
+
+    @migrator.create_model
+    class Tag(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        name = pw.CharField(max_length=255)
+        user_id = pw.CharField(max_length=255)
+        data = pw.TextField(null=True)
+
+        class Meta:
+            table_name = "tag"
+
+    @migrator.create_model
+    class User(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        name = pw.CharField(max_length=255)
+        email = pw.CharField(max_length=255)
+        role = pw.CharField(max_length=255)
+        profile_image_url = pw.TextField()
+        timestamp = pw.BigIntegerField()
 
         class Meta:
             table_name = "user"

backend/apps/web/internal/migrations/004_add_archived.py

@@ -0,0 +1,46 @@
+"""Peewee migrations -- 002_add_local_sharing.py.
+
+Some examples (model - class or model name)::
+
+    > Model = migrator.orm['table_name']            # Return model in current state by name
+    > Model = migrator.ModelClass                   # Return model in current state by name
+
+    > migrator.sql(sql)                             # Run custom SQL
+    > migrator.run(func, *args, **kwargs)           # Run python function with the given args
+    > migrator.create_model(Model)                  # Create a model (could be used as decorator)
+    > migrator.remove_model(model, cascade=True)    # Remove a model
+    > migrator.add_fields(model, **fields)          # Add fields to a model
+    > migrator.change_fields(model, **fields)       # Change fields
+    > migrator.remove_fields(model, *field_names, cascade=True)
+    > migrator.rename_field(model, old_field_name, new_field_name)
+    > migrator.rename_table(model, new_table_name)
+    > migrator.add_index(model, *col_names, unique=False)
+    > migrator.add_not_null(model, *field_names)
+    > migrator.add_default(model, field_name, default)
+    > migrator.add_constraint(model, name, sql)
+    > migrator.drop_index(model, *col_names)
+    > migrator.drop_not_null(model, *field_names)
+    > migrator.drop_constraints(model, *constraints)
+
+"""
+
+from contextlib import suppress
+
+import peewee as pw
+from peewee_migrate import Migrator
+
+
+with suppress(ImportError):
+    import playhouse.postgres_ext as pw_pext
+
+
+def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your migrations here."""
+
+    migrator.add_fields("chat", archived=pw.BooleanField(default=False))
+
+
+def rollback(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your rollback migrations here."""
+
+    migrator.remove_fields("chat", "archived")

backend/apps/web/internal/migrations/005_add_updated_at.py

@@ -0,0 +1,130 @@
+"""Peewee migrations -- 002_add_local_sharing.py.
+
+Some examples (model - class or model name)::
+
+    > Model = migrator.orm['table_name']            # Return model in current state by name
+    > Model = migrator.ModelClass                   # Return model in current state by name
+
+    > migrator.sql(sql)                             # Run custom SQL
+    > migrator.run(func, *args, **kwargs)           # Run python function with the given args
+    > migrator.create_model(Model)                  # Create a model (could be used as decorator)
+    > migrator.remove_model(model, cascade=True)    # Remove a model
+    > migrator.add_fields(model, **fields)          # Add fields to a model
+    > migrator.change_fields(model, **fields)       # Change fields
+    > migrator.remove_fields(model, *field_names, cascade=True)
+    > migrator.rename_field(model, old_field_name, new_field_name)
+    > migrator.rename_table(model, new_table_name)
+    > migrator.add_index(model, *col_names, unique=False)
+    > migrator.add_not_null(model, *field_names)
+    > migrator.add_default(model, field_name, default)
+    > migrator.add_constraint(model, name, sql)
+    > migrator.drop_index(model, *col_names)
+    > migrator.drop_not_null(model, *field_names)
+    > migrator.drop_constraints(model, *constraints)
+
+"""
+
+from contextlib import suppress
+
+import peewee as pw
+from peewee_migrate import Migrator
+
+
+with suppress(ImportError):
+    import playhouse.postgres_ext as pw_pext
+
+
+def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your migrations here."""
+
+    if isinstance(database, pw.SqliteDatabase):
+        migrate_sqlite(migrator, database, fake=fake)
+    else:
+        migrate_external(migrator, database, fake=fake)
+
+
+def migrate_sqlite(migrator: Migrator, database: pw.Database, *, fake=False):
+    # Adding fields created_at and updated_at to the 'chat' table
+    migrator.add_fields(
+        "chat",
+        created_at=pw.DateTimeField(null=True),  # Allow null for transition
+        updated_at=pw.DateTimeField(null=True),  # Allow null for transition
+    )
+
+    # Populate the new fields from an existing 'timestamp' field
+    migrator.sql(
+        "UPDATE chat SET created_at = timestamp, updated_at = timestamp WHERE timestamp IS NOT NULL"
+    )
+
+    # Now that the data has been copied, remove the original 'timestamp' field
+    migrator.remove_fields("chat", "timestamp")
+
+    # Update the fields to be not null now that they are populated
+    migrator.change_fields(
+        "chat",
+        created_at=pw.DateTimeField(null=False),
+        updated_at=pw.DateTimeField(null=False),
+    )
+
+
+def migrate_external(migrator: Migrator, database: pw.Database, *, fake=False):
+    # Adding fields created_at and updated_at to the 'chat' table
+    migrator.add_fields(
+        "chat",
+        created_at=pw.BigIntegerField(null=True),  # Allow null for transition
+        updated_at=pw.BigIntegerField(null=True),  # Allow null for transition
+    )
+
+    # Populate the new fields from an existing 'timestamp' field
+    migrator.sql(
+        "UPDATE chat SET created_at = timestamp, updated_at = timestamp WHERE timestamp IS NOT NULL"
+    )
+
+    # Now that the data has been copied, remove the original 'timestamp' field
+    migrator.remove_fields("chat", "timestamp")
+
+    # Update the fields to be not null now that they are populated
+    migrator.change_fields(
+        "chat",
+        created_at=pw.BigIntegerField(null=False),
+        updated_at=pw.BigIntegerField(null=False),
+    )
+
+
+def rollback(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your rollback migrations here."""
+
+    if isinstance(database, pw.SqliteDatabase):
+        rollback_sqlite(migrator, database, fake=fake)
+    else:
+        rollback_external(migrator, database, fake=fake)
+
+
+def rollback_sqlite(migrator: Migrator, database: pw.Database, *, fake=False):
+    # Recreate the timestamp field initially allowing null values for safe transition
+    migrator.add_fields("chat", timestamp=pw.DateTimeField(null=True))
+
+    # Copy the earliest created_at date back into the new timestamp field
+    # This assumes created_at was originally a copy of timestamp
+    migrator.sql("UPDATE chat SET timestamp = created_at")
+
+    # Remove the created_at and updated_at fields
+    migrator.remove_fields("chat", "created_at", "updated_at")
+
+    # Finally, alter the timestamp field to not allow nulls if that was the original setting
+    migrator.change_fields("chat", timestamp=pw.DateTimeField(null=False))
+
+
+def rollback_external(migrator: Migrator, database: pw.Database, *, fake=False):
+    # Recreate the timestamp field initially allowing null values for safe transition
+    migrator.add_fields("chat", timestamp=pw.BigIntegerField(null=True))
+
+    # Copy the earliest created_at date back into the new timestamp field
+    # This assumes created_at was originally a copy of timestamp
+    migrator.sql("UPDATE chat SET timestamp = created_at")
+
+    # Remove the created_at and updated_at fields
+    migrator.remove_fields("chat", "created_at", "updated_at")
+
+    # Finally, alter the timestamp field to not allow nulls if that was the original setting
+    migrator.change_fields("chat", timestamp=pw.BigIntegerField(null=False))

backend/apps/web/internal/migrations/006_migrate_timestamps_and_charfields.py

@@ -0,0 +1,130 @@
+"""Peewee migrations -- 006_migrate_timestamps_and_charfields.py.
+
+Some examples (model - class or model name)::
+
+    > Model = migrator.orm['table_name']            # Return model in current state by name
+    > Model = migrator.ModelClass                   # Return model in current state by name
+
+    > migrator.sql(sql)                             # Run custom SQL
+    > migrator.run(func, *args, **kwargs)           # Run python function with the given args
+    > migrator.create_model(Model)                  # Create a model (could be used as decorator)
+    > migrator.remove_model(model, cascade=True)    # Remove a model
+    > migrator.add_fields(model, **fields)          # Add fields to a model
+    > migrator.change_fields(model, **fields)       # Change fields
+    > migrator.remove_fields(model, *field_names, cascade=True)
+    > migrator.rename_field(model, old_field_name, new_field_name)
+    > migrator.rename_table(model, new_table_name)
+    > migrator.add_index(model, *col_names, unique=False)
+    > migrator.add_not_null(model, *field_names)
+    > migrator.add_default(model, field_name, default)
+    > migrator.add_constraint(model, name, sql)
+    > migrator.drop_index(model, *col_names)
+    > migrator.drop_not_null(model, *field_names)
+    > migrator.drop_constraints(model, *constraints)
+
+"""
+
+from contextlib import suppress
+
+import peewee as pw
+from peewee_migrate import Migrator
+
+
+with suppress(ImportError):
+    import playhouse.postgres_ext as pw_pext
+
+
+def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your migrations here."""
+
+    # Alter the tables with timestamps
+    migrator.change_fields(
+        "chatidtag",
+        timestamp=pw.BigIntegerField(),
+    )
+    migrator.change_fields(
+        "document",
+        timestamp=pw.BigIntegerField(),
+    )
+    migrator.change_fields(
+        "modelfile",
+        timestamp=pw.BigIntegerField(),
+    )
+    migrator.change_fields(
+        "prompt",
+        timestamp=pw.BigIntegerField(),
+    )
+    migrator.change_fields(
+        "user",
+        timestamp=pw.BigIntegerField(),
+    )
+    # Alter the tables with varchar to text where necessary
+    migrator.change_fields(
+        "auth",
+        password=pw.TextField(),
+    )
+    migrator.change_fields(
+        "chat",
+        title=pw.TextField(),
+    )
+    migrator.change_fields(
+        "document",
+        title=pw.TextField(),
+        filename=pw.TextField(),
+    )
+    migrator.change_fields(
+        "prompt",
+        title=pw.TextField(),
+    )
+    migrator.change_fields(
+        "user",
+        profile_image_url=pw.TextField(),
+    )
+
+
+def rollback(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your rollback migrations here."""
+
+    if isinstance(database, pw.SqliteDatabase):
+        # Alter the tables with timestamps
+        migrator.change_fields(
+            "chatidtag",
+            timestamp=pw.DateField(),
+        )
+        migrator.change_fields(
+            "document",
+            timestamp=pw.DateField(),
+        )
+        migrator.change_fields(
+            "modelfile",
+            timestamp=pw.DateField(),
+        )
+        migrator.change_fields(
+            "prompt",
+            timestamp=pw.DateField(),
+        )
+        migrator.change_fields(
+            "user",
+            timestamp=pw.DateField(),
+        )
+    migrator.change_fields(
+        "auth",
+        password=pw.CharField(max_length=255),
+    )
+    migrator.change_fields(
+        "chat",
+        title=pw.CharField(),
+    )
+    migrator.change_fields(
+        "document",
+        title=pw.CharField(),
+        filename=pw.CharField(),
+    )
+    migrator.change_fields(
+        "prompt",
+        title=pw.CharField(),
+    )
+    migrator.change_fields(
+        "user",
+        profile_image_url=pw.CharField(),
+    )

backend/apps/web/internal/migrations/007_add_user_last_active_at.py

@@ -0,0 +1,79 @@
+"""Peewee migrations -- 002_add_local_sharing.py.
+
+Some examples (model - class or model name)::
+
+    > Model = migrator.orm['table_name']            # Return model in current state by name
+    > Model = migrator.ModelClass                   # Return model in current state by name
+
+    > migrator.sql(sql)                             # Run custom SQL
+    > migrator.run(func, *args, **kwargs)           # Run python function with the given args
+    > migrator.create_model(Model)                  # Create a model (could be used as decorator)
+    > migrator.remove_model(model, cascade=True)    # Remove a model
+    > migrator.add_fields(model, **fields)          # Add fields to a model
+    > migrator.change_fields(model, **fields)       # Change fields
+    > migrator.remove_fields(model, *field_names, cascade=True)
+    > migrator.rename_field(model, old_field_name, new_field_name)
+    > migrator.rename_table(model, new_table_name)
+    > migrator.add_index(model, *col_names, unique=False)
+    > migrator.add_not_null(model, *field_names)
+    > migrator.add_default(model, field_name, default)
+    > migrator.add_constraint(model, name, sql)
+    > migrator.drop_index(model, *col_names)
+    > migrator.drop_not_null(model, *field_names)
+    > migrator.drop_constraints(model, *constraints)
+
+"""
+
+from contextlib import suppress
+
+import peewee as pw
+from peewee_migrate import Migrator
+
+
+with suppress(ImportError):
+    import playhouse.postgres_ext as pw_pext
+
+
+def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your migrations here."""
+
+    # Adding fields created_at and updated_at to the 'user' table
+    migrator.add_fields(
+        "user",
+        created_at=pw.BigIntegerField(null=True),  # Allow null for transition
+        updated_at=pw.BigIntegerField(null=True),  # Allow null for transition
+        last_active_at=pw.BigIntegerField(null=True),  # Allow null for transition
+    )
+
+    # Populate the new fields from an existing 'timestamp' field
+    migrator.sql(
+        'UPDATE "user" SET created_at = timestamp, updated_at = timestamp, last_active_at = timestamp WHERE timestamp IS NOT NULL'
+    )
+
+    # Now that the data has been copied, remove the original 'timestamp' field
+    migrator.remove_fields("user", "timestamp")
+
+    # Update the fields to be not null now that they are populated
+    migrator.change_fields(
+        "user",
+        created_at=pw.BigIntegerField(null=False),
+        updated_at=pw.BigIntegerField(null=False),
+        last_active_at=pw.BigIntegerField(null=False),
+    )
+
+
+def rollback(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your rollback migrations here."""
+
+    # Recreate the timestamp field initially allowing null values for safe transition
+    migrator.add_fields("user", timestamp=pw.BigIntegerField(null=True))
+
+    # Copy the earliest created_at date back into the new timestamp field
+    # This assumes created_at was originally a copy of timestamp
+    migrator.sql('UPDATE "user" SET timestamp = created_at')
+
+    # Remove the created_at and updated_at fields
+    migrator.remove_fields("user", "created_at", "updated_at", "last_active_at")
+
+    # Finally, alter the timestamp field to not allow nulls if that was the original setting
+    migrator.change_fields("user", timestamp=pw.BigIntegerField(null=False))

backend/apps/web/internal/migrations/008_add_memory.py

@@ -0,0 +1,53 @@
+"""Peewee migrations -- 002_add_local_sharing.py.
+
+Some examples (model - class or model name)::
+
+    > Model = migrator.orm['table_name']            # Return model in current state by name
+    > Model = migrator.ModelClass                   # Return model in current state by name
+
+    > migrator.sql(sql)                             # Run custom SQL
+    > migrator.run(func, *args, **kwargs)           # Run python function with the given args
+    > migrator.create_model(Model)                  # Create a model (could be used as decorator)
+    > migrator.remove_model(model, cascade=True)    # Remove a model
+    > migrator.add_fields(model, **fields)          # Add fields to a model
+    > migrator.change_fields(model, **fields)       # Change fields
+    > migrator.remove_fields(model, *field_names, cascade=True)
+    > migrator.rename_field(model, old_field_name, new_field_name)
+    > migrator.rename_table(model, new_table_name)
+    > migrator.add_index(model, *col_names, unique=False)
+    > migrator.add_not_null(model, *field_names)
+    > migrator.add_default(model, field_name, default)
+    > migrator.add_constraint(model, name, sql)
+    > migrator.drop_index(model, *col_names)
+    > migrator.drop_not_null(model, *field_names)
+    > migrator.drop_constraints(model, *constraints)
+
+"""
+
+from contextlib import suppress
+
+import peewee as pw
+from peewee_migrate import Migrator
+
+
+with suppress(ImportError):
+    import playhouse.postgres_ext as pw_pext
+
+
+def migrate(migrator: Migrator, database: pw.Database, *, fake=False):
+    @migrator.create_model
+    class Memory(pw.Model):
+        id = pw.CharField(max_length=255, unique=True)
+        user_id = pw.CharField(max_length=255)
+        content = pw.TextField(null=False)
+        updated_at = pw.BigIntegerField(null=False)
+        created_at = pw.BigIntegerField(null=False)
+
+        class Meta:
+            table_name = "memory"
+
+
+def rollback(migrator: Migrator, database: pw.Database, *, fake=False):
+    """Write your rollback migrations here."""
+
+    migrator.remove_model("memory")

backend/apps/web/main.py

@@ -9,6 +9,7 @@ from apps.web.routers import (
     modelfiles,
     prompts,
     configs,
+    memories,
     utils,
 )
 from config import (
@@ -21,22 +22,27 @@ from config import (
     USER_PERMISSIONS,
     WEBHOOK_URL,
     WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
+    JWT_EXPIRES_IN,
+    AppConfig,
 )
 
 app = FastAPI()
 
 origins = ["*"]
 
-app.state.ENABLE_SIGNUP = ENABLE_SIGNUP
-app.state.JWT_EXPIRES_IN = "-1"
+app.state.config = AppConfig()
 
-app.state.DEFAULT_MODELS = DEFAULT_MODELS
-app.state.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS
-app.state.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE
-app.state.USER_PERMISSIONS = USER_PERMISSIONS
-app.state.WEBHOOK_URL = WEBHOOK_URL
+app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP
+app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN
+
+app.state.config.DEFAULT_MODELS = DEFAULT_MODELS
+app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS
+app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE
+app.state.config.USER_PERMISSIONS = USER_PERMISSIONS
+app.state.config.WEBHOOK_URL = WEBHOOK_URL
 app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER
 
+
 app.add_middleware(
     CORSMiddleware,
     allow_origins=origins,
@@ -48,9 +54,12 @@ app.add_middleware(
 app.include_router(auths.router, prefix="/auths", tags=["auths"])
 app.include_router(users.router, prefix="/users", tags=["users"])
 app.include_router(chats.router, prefix="/chats", tags=["chats"])
+
 app.include_router(documents.router, prefix="/documents", tags=["documents"])
 app.include_router(modelfiles.router, prefix="/modelfiles", tags=["modelfiles"])
 app.include_router(prompts.router, prefix="/prompts", tags=["prompts"])
+app.include_router(memories.router, prefix="/memories", tags=["memories"])
+
 
 app.include_router(configs.router, prefix="/configs", tags=["configs"])
 app.include_router(utils.router, prefix="/utils", tags=["utils"])
@@ -61,6 +70,6 @@ async def get_status():
     return {
         "status": True,
         "auth": WEBUI_AUTH,
-        "default_models": app.state.DEFAULT_MODELS,
-        "default_prompt_suggestions": app.state.DEFAULT_PROMPT_SUGGESTIONS,
+        "default_models": app.state.config.DEFAULT_MODELS,
+        "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS,
     }

backend/apps/web/models/auths.py

@@ -23,7 +23,7 @@ log.setLevel(SRC_LOG_LEVELS["MODELS"])
 class Auth(Model):
     id = CharField(unique=True)
     email = CharField()
-    password = CharField()
+    password = TextField()
     active = BooleanField()
 
     class Meta:
@@ -89,6 +89,10 @@ class SignupForm(BaseModel):
     profile_image_url: Optional[str] = "/user.png"
 
 
+class AddUserForm(SignupForm):
+    role: Optional[str] = "pending"
+
+
 class AuthsTable:
     def __init__(self, db):
         self.db = db

backend/apps/web/models/chats.py

@@ -17,10 +17,14 @@ from apps.web.internal.db import DB
 class Chat(Model):
     id = CharField(unique=True)
     user_id = CharField()
-    title = CharField()
+    title = TextField()
     chat = TextField()  # Save Chat JSON as Text
-    timestamp = DateField()
+
+    created_at = BigIntegerField()
+    updated_at = BigIntegerField()
+
     share_id = CharField(null=True, unique=True)
+    archived = BooleanField(default=False)
 
     class Meta:
         database = DB
@@ -31,8 +35,12 @@ class ChatModel(BaseModel):
     user_id: str
     title: str
     chat: str
-    timestamp: int  # timestamp in epoch
+
+    created_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+
     share_id: Optional[str] = None
+    archived: bool = False
 
 
 ####################
@@ -53,13 +61,17 @@ class ChatResponse(BaseModel):
     user_id: str
     title: str
     chat: dict
-    timestamp: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
     share_id: Optional[str] = None  # id of the chat to be shared
+    archived: bool
 
 
 class ChatTitleIdResponse(BaseModel):
     id: str
     title: str
+    updated_at: int
+    created_at: int
 
 
 class ChatTable:
@@ -77,7 +89,8 @@ class ChatTable:
                     form_data.chat["title"] if "title" in form_data.chat else "New Chat"
                 ),
                 "chat": json.dumps(form_data.chat),
-                "timestamp": int(time.time()),
+                "created_at": int(time.time()),
+                "updated_at": int(time.time()),
             }
         )
 
@@ -89,7 +102,7 @@ class ChatTable:
             query = Chat.update(
                 chat=json.dumps(chat),
                 title=chat["title"] if "title" in chat else "New Chat",
-                timestamp=int(time.time()),
+                updated_at=int(time.time()),
             ).where(Chat.id == id)
             query.execute()
 
@@ -111,7 +124,8 @@ class ChatTable:
                 "user_id": f"shared-{chat_id}",
                 "title": chat.title,
                 "chat": chat.chat,
-                "timestamp": int(time.time()),
+                "created_at": chat.created_at,
+                "updated_at": int(time.time()),
             }
         )
         shared_result = Chat.create(**shared_chat.model_dump())
@@ -163,40 +177,55 @@ class ChatTable:
         except:
             return None
 
-    def get_chat_lists_by_user_id(
+    def toggle_chat_archive_by_id(self, id: str) -> Optional[ChatModel]:
+        try:
+            chat = self.get_chat_by_id(id)
+            query = Chat.update(
+                archived=(not chat.archived),
+            ).where(Chat.id == id)
+
+            query.execute()
+
+            chat = Chat.get(Chat.id == id)
+            return ChatModel(**model_to_dict(chat))
+        except:
+            return None
+
+    def get_archived_chat_list_by_user_id(
         self, user_id: str, skip: int = 0, limit: int = 50
     ) -> List[ChatModel]:
         return [
             ChatModel(**model_to_dict(chat))
             for chat in Chat.select()
+            .where(Chat.archived == True)
             .where(Chat.user_id == user_id)
-            .order_by(Chat.timestamp.desc())
+            .order_by(Chat.updated_at.desc())
             # .limit(limit)
             # .offset(skip)
         ]
 
-    def get_chat_lists_by_chat_ids(
+    def get_chat_list_by_user_id(
+        self, user_id: str, skip: int = 0, limit: int = 50
+    ) -> List[ChatModel]:
+        return [
+            ChatModel(**model_to_dict(chat))
+            for chat in Chat.select()
+            .where(Chat.archived == False)
+            .where(Chat.user_id == user_id)
+            .order_by(Chat.updated_at.desc())
+            # .limit(limit)
+            # .offset(skip)
+        ]
+
+    def get_chat_list_by_chat_ids(
         self, chat_ids: List[str], skip: int = 0, limit: int = 50
     ) -> List[ChatModel]:
         return [
             ChatModel(**model_to_dict(chat))
             for chat in Chat.select()
+            .where(Chat.archived == False)
             .where(Chat.id.in_(chat_ids))
-            .order_by(Chat.timestamp.desc())
-        ]
-
-    def get_all_chats(self) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select().order_by(Chat.timestamp.desc())
-        ]
-
-    def get_all_chats_by_user_id(self, user_id: str) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select()
-            .where(Chat.user_id == user_id)
-            .order_by(Chat.timestamp.desc())
+            .order_by(Chat.updated_at.desc())
         ]
 
     def get_chat_by_id(self, id: str) -> Optional[ChatModel]:
@@ -228,9 +257,28 @@ class ChatTable:
     def get_chats(self, skip: int = 0, limit: int = 50) -> List[ChatModel]:
         return [
             ChatModel(**model_to_dict(chat))
-            for chat in Chat.select().limit(limit).offset(skip)
+            for chat in Chat.select().order_by(Chat.updated_at.desc())
+            # .limit(limit).offset(skip)
         ]
 
+    def get_chats_by_user_id(self, user_id: str) -> List[ChatModel]:
+        return [
+            ChatModel(**model_to_dict(chat))
+            for chat in Chat.select()
+            .where(Chat.user_id == user_id)
+            .order_by(Chat.updated_at.desc())
+            # .limit(limit).offset(skip)
+        ]
+
+    def delete_chat_by_id(self, id: str) -> bool:
+        try:
+            query = Chat.delete().where((Chat.id == id))
+            query.execute()  # Remove the rows, return number of rows removed.
+
+            return True and self.delete_shared_chat_by_chat_id(id)
+        except:
+            return False
+
     def delete_chat_by_id_and_user_id(self, id: str, user_id: str) -> bool:
         try:
             query = Chat.delete().where((Chat.id == id) & (Chat.user_id == user_id))

backend/apps/web/models/documents.py

@@ -25,11 +25,11 @@ log.setLevel(SRC_LOG_LEVELS["MODELS"])
 class Document(Model):
     collection_name = CharField(unique=True)
     name = CharField(unique=True)
-    title = CharField()
-    filename = CharField()
+    title = TextField()
+    filename = TextField()
     content = TextField(null=True)
     user_id = CharField()
-    timestamp = DateField()
+    timestamp = BigIntegerField()
 
     class Meta:
         database = DB

backend/apps/web/models/memories.py

@@ -0,0 +1,118 @@
+from pydantic import BaseModel
+from peewee import *
+from playhouse.shortcuts import model_to_dict
+from typing import List, Union, Optional
+
+from apps.web.internal.db import DB
+from apps.web.models.chats import Chats
+
+import time
+import uuid
+
+####################
+# Memory DB Schema
+####################
+
+
+class Memory(Model):
+    id = CharField(unique=True)
+    user_id = CharField()
+    content = TextField()
+    updated_at = BigIntegerField()
+    created_at = BigIntegerField()
+
+    class Meta:
+        database = DB
+
+
+class MemoryModel(BaseModel):
+    id: str
+    user_id: str
+    content: str
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+
+####################
+# Forms
+####################
+
+
+class MemoriesTable:
+    def __init__(self, db):
+        self.db = db
+        self.db.create_tables([Memory])
+
+    def insert_new_memory(
+        self,
+        user_id: str,
+        content: str,
+    ) -> Optional[MemoryModel]:
+        id = str(uuid.uuid4())
+
+        memory = MemoryModel(
+            **{
+                "id": id,
+                "user_id": user_id,
+                "content": content,
+                "created_at": int(time.time()),
+                "updated_at": int(time.time()),
+            }
+        )
+        result = Memory.create(**memory.model_dump())
+        if result:
+            return memory
+        else:
+            return None
+
+    def get_memories(self) -> List[MemoryModel]:
+        try:
+            memories = Memory.select()
+            return [MemoryModel(**model_to_dict(memory)) for memory in memories]
+        except:
+            return None
+
+    def get_memories_by_user_id(self, user_id: str) -> List[MemoryModel]:
+        try:
+            memories = Memory.select().where(Memory.user_id == user_id)
+            return [MemoryModel(**model_to_dict(memory)) for memory in memories]
+        except:
+            return None
+
+    def get_memory_by_id(self, id) -> Optional[MemoryModel]:
+        try:
+            memory = Memory.get(Memory.id == id)
+            return MemoryModel(**model_to_dict(memory))
+        except:
+            return None
+
+    def delete_memory_by_id(self, id: str) -> bool:
+        try:
+            query = Memory.delete().where(Memory.id == id)
+            query.execute()  # Remove the rows, return number of rows removed.
+
+            return True
+
+        except:
+            return False
+
+    def delete_memories_by_user_id(self, user_id: str) -> bool:
+        try:
+            query = Memory.delete().where(Memory.user_id == user_id)
+            query.execute()
+
+            return True
+        except:
+            return False
+
+    def delete_memory_by_id_and_user_id(self, id: str, user_id: str) -> bool:
+        try:
+            query = Memory.delete().where(Memory.id == id, Memory.user_id == user_id)
+            query.execute()
+
+            return True
+        except:
+            return False
+
+
+Memories = MemoriesTable(DB)

backend/apps/web/models/modelfiles.py

@@ -20,7 +20,7 @@ class Modelfile(Model):
     tag_name = CharField(unique=True)
     user_id = CharField()
     modelfile = TextField()
-    timestamp = DateField()
+    timestamp = BigIntegerField()
 
     class Meta:
         database = DB

backend/apps/web/models/prompts.py

@@ -19,9 +19,9 @@ import json
 class Prompt(Model):
     command = CharField(unique=True)
     user_id = CharField()
-    title = CharField()
+    title = TextField()
     content = TextField()
-    timestamp = DateField()
+    timestamp = BigIntegerField()
 
     class Meta:
         database = DB

backend/apps/web/models/tags.py

@@ -35,7 +35,7 @@ class ChatIdTag(Model):
     tag_name = CharField()
     chat_id = CharField()
     user_id = CharField()
-    timestamp = DateField()
+    timestamp = BigIntegerField()
 
     class Meta:
         database = DB
@@ -136,7 +136,9 @@ class TagTable:
 
         return [
             TagModel(**model_to_dict(tag))
-            for tag in Tag.select().where(Tag.name.in_(tag_names))
+            for tag in Tag.select()
+            .where(Tag.user_id == user_id)
+            .where(Tag.name.in_(tag_names))
         ]
 
     def get_tags_by_chat_id_and_user_id(
@@ -151,7 +153,9 @@ class TagTable:
 
         return [
             TagModel(**model_to_dict(tag))
-            for tag in Tag.select().where(Tag.name.in_(tag_names))
+            for tag in Tag.select()
+            .where(Tag.user_id == user_id)
+            .where(Tag.name.in_(tag_names))
         ]
 
     def get_chat_ids_by_tag_name_and_user_id(

backend/apps/web/models/users.py

@@ -18,8 +18,12 @@ class User(Model):
     name = CharField()
     email = CharField()
     role = CharField()
-    profile_image_url = CharField()
-    timestamp = DateField()
+    profile_image_url = TextField()
+
+    last_active_at = BigIntegerField()
+    updated_at = BigIntegerField()
+    created_at = BigIntegerField()
+
     api_key = CharField(null=True, unique=True)
 
     class Meta:
@@ -32,7 +36,11 @@ class UserModel(BaseModel):
     email: str
     role: str = "pending"
     profile_image_url: str
-    timestamp: int  # timestamp in epoch
+
+    last_active_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
     api_key: Optional[str] = None
 
 
@@ -73,7 +81,9 @@ class UsersTable:
                 "email": email,
                 "role": role,
                 "profile_image_url": profile_image_url,
-                "timestamp": int(time.time()),
+                "last_active_at": int(time.time()),
+                "created_at": int(time.time()),
+                "updated_at": int(time.time()),
             }
         )
         result = User.create(**user.model_dump())
@@ -113,6 +123,13 @@ class UsersTable:
     def get_num_users(self) -> Optional[int]:
         return User.select().count()
 
+    def get_first_user(self) -> UserModel:
+        try:
+            user = User.select().order_by(User.created_at).first()
+            return UserModel(**model_to_dict(user))
+        except:
+            return None
+
     def update_user_role_by_id(self, id: str, role: str) -> Optional[UserModel]:
         try:
             query = User.update(role=role).where(User.id == id)
@@ -137,6 +154,16 @@ class UsersTable:
         except:
             return None
 
+    def update_user_last_active_by_id(self, id: str) -> Optional[UserModel]:
+        try:
+            query = User.update(last_active_at=int(time.time())).where(User.id == id)
+            query.execute()
+
+            user = User.get(User.id == id)
+            return UserModel(**model_to_dict(user))
+        except:
+            return None
+
     def update_user_by_id(self, id: str, updated: dict) -> Optional[UserModel]:
         try:
             query = User.update(**updated).where(User.id == id)

backend/apps/web/routers/auths.py

@@ -1,14 +1,19 @@
-from fastapi import Request
+import logging
+
+from fastapi import Request, UploadFile, File
 from fastapi import Depends, HTTPException, status
 
 from fastapi import APIRouter
 from pydantic import BaseModel
 import re
 import uuid
+import csv
+
 
 from apps.web.models.auths import (
     SigninForm,
     SignupForm,
+    AddUserForm,
     UpdateProfileForm,
     UpdatePasswordForm,
     UserResponse,
@@ -28,7 +33,7 @@ from utils.utils import (
 from utils.misc import parse_duration, validate_email_format
 from utils.webhook import post_webhook
 from constants import ERROR_MESSAGES, WEBHOOK_MESSAGES
-from config import WEBUI_AUTH_TRUSTED_EMAIL_HEADER
+from config import WEBUI_AUTH, WEBUI_AUTH_TRUSTED_EMAIL_HEADER
 
 router = APIRouter()
 
@@ -113,13 +118,29 @@ async def signin(request: Request, form_data: SigninForm):
                 ),
             )
         user = Auths.authenticate_user_by_trusted_header(trusted_email)
+    elif WEBUI_AUTH == False:
+        admin_email = "admin@localhost"
+        admin_password = "admin"
+
+        if Users.get_user_by_email(admin_email.lower()):
+            user = Auths.authenticate_user(admin_email.lower(), admin_password)
+        else:
+            if Users.get_num_users() != 0:
+                raise HTTPException(400, detail=ERROR_MESSAGES.EXISTING_USERS)
+
+            await signup(
+                request,
+                SignupForm(email=admin_email, password=admin_password, name="User"),
+            )
+
+            user = Auths.authenticate_user(admin_email.lower(), admin_password)
     else:
         user = Auths.authenticate_user(form_data.email.lower(), form_data.password)
 
     if user:
         token = create_token(
             data={"id": user.id},
-            expires_delta=parse_duration(request.app.state.JWT_EXPIRES_IN),
+            expires_delta=parse_duration(request.app.state.config.JWT_EXPIRES_IN),
         )
 
         return {
@@ -142,7 +163,7 @@ async def signin(request: Request, form_data: SigninForm):
 
 @router.post("/signup", response_model=SigninResponse)
 async def signup(request: Request, form_data: SignupForm):
-    if not request.app.state.ENABLE_SIGNUP:
+    if not request.app.state.config.ENABLE_SIGNUP and WEBUI_AUTH:
         raise HTTPException(
             status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.ACCESS_PROHIBITED
         )
@@ -159,7 +180,7 @@ async def signup(request: Request, form_data: SignupForm):
         role = (
             "admin"
             if Users.get_num_users() == 0
-            else request.app.state.DEFAULT_USER_ROLE
+            else request.app.state.config.DEFAULT_USER_ROLE
         )
         hashed = get_password_hash(form_data.password)
         user = Auths.insert_new_auth(
@@ -173,13 +194,13 @@ async def signup(request: Request, form_data: SignupForm):
         if user:
             token = create_token(
                 data={"id": user.id},
-                expires_delta=parse_duration(request.app.state.JWT_EXPIRES_IN),
+                expires_delta=parse_duration(request.app.state.config.JWT_EXPIRES_IN),
             )
             # response.set_cookie(key='token', value=token, httponly=True)
 
-            if request.app.state.WEBHOOK_URL:
+            if request.app.state.config.WEBHOOK_URL:
                 post_webhook(
-                    request.app.state.WEBHOOK_URL,
+                    request.app.state.config.WEBHOOK_URL,
                     WEBHOOK_MESSAGES.USER_SIGNUP(user.name),
                     {
                         "action": "signup",
@@ -203,6 +224,51 @@ async def signup(request: Request, form_data: SignupForm):
         raise HTTPException(500, detail=ERROR_MESSAGES.DEFAULT(err))
 
 
+############################
+# AddUser
+############################
+
+
+@router.post("/add", response_model=SigninResponse)
+async def add_user(form_data: AddUserForm, user=Depends(get_admin_user)):
+
+    if not validate_email_format(form_data.email.lower()):
+        raise HTTPException(
+            status.HTTP_400_BAD_REQUEST, detail=ERROR_MESSAGES.INVALID_EMAIL_FORMAT
+        )
+
+    if Users.get_user_by_email(form_data.email.lower()):
+        raise HTTPException(400, detail=ERROR_MESSAGES.EMAIL_TAKEN)
+
+    try:
+
+        print(form_data)
+        hashed = get_password_hash(form_data.password)
+        user = Auths.insert_new_auth(
+            form_data.email.lower(),
+            hashed,
+            form_data.name,
+            form_data.profile_image_url,
+            form_data.role,
+        )
+
+        if user:
+            token = create_token(data={"id": user.id})
+            return {
+                "token": token,
+                "token_type": "Bearer",
+                "id": user.id,
+                "email": user.email,
+                "name": user.name,
+                "role": user.role,
+                "profile_image_url": user.profile_image_url,
+            }
+        else:
+            raise HTTPException(500, detail=ERROR_MESSAGES.CREATE_USER_ERROR)
+    except Exception as err:
+        raise HTTPException(500, detail=ERROR_MESSAGES.DEFAULT(err))
+
+
 ############################
 # ToggleSignUp
 ############################
@@ -210,13 +276,13 @@ async def signup(request: Request, form_data: SignupForm):
 
 @router.get("/signup/enabled", response_model=bool)
 async def get_sign_up_status(request: Request, user=Depends(get_admin_user)):
-    return request.app.state.ENABLE_SIGNUP
+    return request.app.state.config.ENABLE_SIGNUP
 
 
 @router.get("/signup/enabled/toggle", response_model=bool)
 async def toggle_sign_up(request: Request, user=Depends(get_admin_user)):
-    request.app.state.ENABLE_SIGNUP = not request.app.state.ENABLE_SIGNUP
-    return request.app.state.ENABLE_SIGNUP
+    request.app.state.config.ENABLE_SIGNUP = not request.app.state.config.ENABLE_SIGNUP
+    return request.app.state.config.ENABLE_SIGNUP
 
 
 ############################
@@ -226,7 +292,7 @@ async def toggle_sign_up(request: Request, user=Depends(get_admin_user)):
 
 @router.get("/signup/user/role")
 async def get_default_user_role(request: Request, user=Depends(get_admin_user)):
-    return request.app.state.DEFAULT_USER_ROLE
+    return request.app.state.config.DEFAULT_USER_ROLE
 
 
 class UpdateRoleForm(BaseModel):
@@ -238,8 +304,8 @@ async def update_default_user_role(
     request: Request, form_data: UpdateRoleForm, user=Depends(get_admin_user)
 ):
     if form_data.role in ["pending", "user", "admin"]:
-        request.app.state.DEFAULT_USER_ROLE = form_data.role
-    return request.app.state.DEFAULT_USER_ROLE
+        request.app.state.config.DEFAULT_USER_ROLE = form_data.role
+    return request.app.state.config.DEFAULT_USER_ROLE
 
 
 ############################
@@ -249,7 +315,7 @@ async def update_default_user_role(
 
 @router.get("/token/expires")
 async def get_token_expires_duration(request: Request, user=Depends(get_admin_user)):
-    return request.app.state.JWT_EXPIRES_IN
+    return request.app.state.config.JWT_EXPIRES_IN
 
 
 class UpdateJWTExpiresDurationForm(BaseModel):
@@ -266,10 +332,10 @@ async def update_token_expires_duration(
 
     # Check if the input string matches the pattern
     if re.match(pattern, form_data.duration):
-        request.app.state.JWT_EXPIRES_IN = form_data.duration
-        return request.app.state.JWT_EXPIRES_IN
+        request.app.state.config.JWT_EXPIRES_IN = form_data.duration
+        return request.app.state.config.JWT_EXPIRES_IN
     else:
-        return request.app.state.JWT_EXPIRES_IN
+        return request.app.state.config.JWT_EXPIRES_IN
 
 
 ############################

backend/apps/web/routers/chats.py

@@ -28,7 +28,7 @@ from apps.web.models.tags import (
 
 from constants import ERROR_MESSAGES
 
-from config import SRC_LOG_LEVELS
+from config import SRC_LOG_LEVELS, ENABLE_ADMIN_EXPORT
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -36,27 +36,98 @@ log.setLevel(SRC_LOG_LEVELS["MODELS"])
 router = APIRouter()
 
 ############################
-# GetChats
+# GetChatList
 ############################
 
 
 @router.get("/", response_model=List[ChatTitleIdResponse])
-async def get_user_chats(
+@router.get("/list", response_model=List[ChatTitleIdResponse])
+async def get_session_user_chat_list(
     user=Depends(get_current_user), skip: int = 0, limit: int = 50
 ):
-    return Chats.get_chat_lists_by_user_id(user.id, skip, limit)
+    return Chats.get_chat_list_by_user_id(user.id, skip, limit)
 
 
 ############################
-# GetAllChats
+# DeleteAllChats
+############################
+
+
+@router.delete("/", response_model=bool)
+async def delete_all_user_chats(request: Request, user=Depends(get_current_user)):
+
+    if (
+        user.role == "user"
+        and not request.app.state.config.USER_PERMISSIONS["chat"]["deletion"]
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+
+    result = Chats.delete_chats_by_user_id(user.id)
+    return result
+
+
+############################
+# GetUserChatList
+############################
+
+
+@router.get("/list/user/{user_id}", response_model=List[ChatTitleIdResponse])
+async def get_user_chat_list_by_user_id(
+    user_id: str, user=Depends(get_admin_user), skip: int = 0, limit: int = 50
+):
+    return Chats.get_chat_list_by_user_id(user_id, skip, limit)
+
+
+############################
+# GetArchivedChats
+############################
+
+
+@router.get("/archived", response_model=List[ChatTitleIdResponse])
+async def get_archived_session_user_chat_list(
+    user=Depends(get_current_user), skip: int = 0, limit: int = 50
+):
+    return Chats.get_archived_chat_list_by_user_id(user.id, skip, limit)
+
+
+############################
+# GetSharedChatById
+############################
+
+
+@router.get("/share/{share_id}", response_model=Optional[ChatResponse])
+async def get_shared_chat_by_id(share_id: str, user=Depends(get_current_user)):
+    if user.role == "pending":
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
+        )
+
+    if user.role == "user":
+        chat = Chats.get_chat_by_share_id(share_id)
+    elif user.role == "admin":
+        chat = Chats.get_chat_by_id(share_id)
+
+    if chat:
+        return ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
+        )
+
+
+############################
+# GetChats
 ############################
 
 
 @router.get("/all", response_model=List[ChatResponse])
-async def get_all_user_chats(user=Depends(get_current_user)):
+async def get_user_chats(user=Depends(get_current_user)):
     return [
         ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
-        for chat in Chats.get_all_chats_by_user_id(user.id)
+        for chat in Chats.get_chats_by_user_id(user.id)
     ]
 
 
@@ -67,9 +138,14 @@ async def get_all_user_chats(user=Depends(get_current_user)):
 
 @router.get("/all/db", response_model=List[ChatResponse])
 async def get_all_user_chats_in_db(user=Depends(get_admin_user)):
+    if not ENABLE_ADMIN_EXPORT:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
     return [
         ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
-        for chat in Chats.get_all_chats()
+        for chat in Chats.get_chats()
     ]
 
 
@@ -90,6 +166,38 @@ async def create_new_chat(form_data: ChatForm, user=Depends(get_current_user)):
         )
 
 
+############################
+# GetChatsByTags
+############################
+
+
+class TagNameForm(BaseModel):
+    name: str
+    skip: Optional[int] = 0
+    limit: Optional[int] = 50
+
+
+@router.post("/tags", response_model=List[ChatTitleIdResponse])
+async def get_user_chat_list_by_tag_name(
+    form_data: TagNameForm, user=Depends(get_current_user)
+):
+
+    print(form_data)
+    chat_ids = [
+        chat_id_tag.chat_id
+        for chat_id_tag in Tags.get_chat_ids_by_tag_name_and_user_id(
+            form_data.name, user.id
+        )
+    ]
+
+    chats = Chats.get_chat_list_by_chat_ids(chat_ids, form_data.skip, form_data.limit)
+
+    if len(chats) == 0:
+        Tags.delete_tag_by_tag_name_and_user_id(form_data.name, user.id)
+
+    return chats
+
+
 ############################
 # GetAllTags
 ############################
@@ -107,28 +215,6 @@ async def get_all_tags(user=Depends(get_current_user)):
         )
 
 
-############################
-# GetChatsByTags
-############################
-
-
-@router.get("/tags/tag/{tag_name}", response_model=List[ChatTitleIdResponse])
-async def get_user_chats_by_tag_name(
-    tag_name: str, user=Depends(get_current_user), skip: int = 0, limit: int = 50
-):
-    chat_ids = [
-        chat_id_tag.chat_id
-        for chat_id_tag in Tags.get_chat_ids_by_tag_name_and_user_id(tag_name, user.id)
-    ]
-
-    chats = Chats.get_chat_lists_by_chat_ids(chat_ids, skip, limit)
-
-    if len(chats) == 0:
-        Tags.delete_tag_by_tag_name_and_user_id(tag_name, user.id)
-
-    return chats
-
-
 ############################
 # GetChatById
 ############################
@@ -176,17 +262,35 @@ async def update_chat_by_id(
 @router.delete("/{id}", response_model=bool)
 async def delete_chat_by_id(request: Request, id: str, user=Depends(get_current_user)):
 
-    if (
-        user.role == "user"
-        and not request.app.state.USER_PERMISSIONS["chat"]["deletion"]
-    ):
-        raise HTTPException(
-            status_code=status.HTTP_401_UNAUTHORIZED,
-            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
-        )
+    if user.role == "admin":
+        result = Chats.delete_chat_by_id(id)
+        return result
+    else:
+        if not request.app.state.config.USER_PERMISSIONS["chat"]["deletion"]:
+            raise HTTPException(
+                status_code=status.HTTP_401_UNAUTHORIZED,
+                detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+            )
 
-    result = Chats.delete_chat_by_id_and_user_id(id, user.id)
-    return result
+        result = Chats.delete_chat_by_id_and_user_id(id, user.id)
+        return result
+
+
+############################
+# ArchiveChat
+############################
+
+
+@router.get("/{id}/archive", response_model=Optional[ChatResponse])
+async def archive_chat_by_id(id: str, user=Depends(get_current_user)):
+    chat = Chats.get_chat_by_id_and_user_id(id, user.id)
+    if chat:
+        chat = Chats.toggle_chat_archive_by_id(id)
+        return ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.DEFAULT()
+        )
 
 
 ############################
@@ -244,31 +348,6 @@ async def delete_shared_chat_by_id(id: str, user=Depends(get_current_user)):
         )
 
 
-############################
-# GetSharedChatById
-############################
-
-
-@router.get("/share/{share_id}", response_model=Optional[ChatResponse])
-async def get_shared_chat_by_id(share_id: str, user=Depends(get_current_user)):
-    if user.role == "pending":
-        raise HTTPException(
-            status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
-        )
-
-    if user.role == "user":
-        chat = Chats.get_chat_by_share_id(share_id)
-    elif user.role == "admin":
-        chat = Chats.get_chat_by_id(share_id)
-
-    if chat:
-        return ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
-    else:
-        raise HTTPException(
-            status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
-        )
-
-
 ############################
 # GetChatTagsById
 ############################
@@ -349,24 +428,3 @@ async def delete_all_chat_tags_by_id(id: str, user=Depends(get_current_user)):
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
         )
-
-
-############################
-# DeleteAllChats
-############################
-
-
-@router.delete("/", response_model=bool)
-async def delete_all_user_chats(request: Request, user=Depends(get_current_user)):
-
-    if (
-        user.role == "user"
-        and not request.app.state.USER_PERMISSIONS["chat"]["deletion"]
-    ):
-        raise HTTPException(
-            status_code=status.HTTP_401_UNAUTHORIZED,
-            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
-        )
-
-    result = Chats.delete_chats_by_user_id(user.id)
-    return result

backend/apps/web/routers/configs.py

@@ -44,8 +44,8 @@ class SetDefaultSuggestionsForm(BaseModel):
 async def set_global_default_models(
     request: Request, form_data: SetDefaultModelsForm, user=Depends(get_admin_user)
 ):
-    request.app.state.DEFAULT_MODELS = form_data.models
-    return request.app.state.DEFAULT_MODELS
+    request.app.state.config.DEFAULT_MODELS = form_data.models
+    return request.app.state.config.DEFAULT_MODELS
 
 
 @router.post("/default/suggestions", response_model=List[PromptSuggestion])
@@ -55,5 +55,5 @@ async def set_global_default_suggestions(
     user=Depends(get_admin_user),
 ):
     data = form_data.model_dump()
-    request.app.state.DEFAULT_PROMPT_SUGGESTIONS = data["suggestions"]
-    return request.app.state.DEFAULT_PROMPT_SUGGESTIONS
+    request.app.state.config.DEFAULT_PROMPT_SUGGESTIONS = data["suggestions"]
+    return request.app.state.config.DEFAULT_PROMPT_SUGGESTIONS

backend/apps/web/routers/memories.py

@@ -0,0 +1,145 @@
+from fastapi import Response, Request
+from fastapi import Depends, FastAPI, HTTPException, status
+from datetime import datetime, timedelta
+from typing import List, Union, Optional
+
+from fastapi import APIRouter
+from pydantic import BaseModel
+import logging
+
+from apps.web.models.memories import Memories, MemoryModel
+
+from utils.utils import get_verified_user
+from constants import ERROR_MESSAGES
+
+from config import SRC_LOG_LEVELS, CHROMA_CLIENT
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+router = APIRouter()
+
+
+@router.get("/ef")
+async def get_embeddings(request: Request):
+    return {"result": request.app.state.EMBEDDING_FUNCTION("hello world")}
+
+
+############################
+# GetMemories
+############################
+
+
+@router.get("/", response_model=List[MemoryModel])
+async def get_memories(user=Depends(get_verified_user)):
+    return Memories.get_memories_by_user_id(user.id)
+
+
+############################
+# AddMemory
+############################
+
+
+class AddMemoryForm(BaseModel):
+    content: str
+
+
+@router.post("/add", response_model=Optional[MemoryModel])
+async def add_memory(
+    request: Request, form_data: AddMemoryForm, user=Depends(get_verified_user)
+):
+    memory = Memories.insert_new_memory(user.id, form_data.content)
+    memory_embedding = request.app.state.EMBEDDING_FUNCTION(memory.content)
+
+    collection = CHROMA_CLIENT.get_or_create_collection(name=f"user-memory-{user.id}")
+    collection.upsert(
+        documents=[memory.content],
+        ids=[memory.id],
+        embeddings=[memory_embedding],
+        metadatas=[{"created_at": memory.created_at}],
+    )
+
+    return memory
+
+
+############################
+# QueryMemory
+############################
+
+
+class QueryMemoryForm(BaseModel):
+    content: str
+
+
+@router.post("/query")
+async def query_memory(
+    request: Request, form_data: QueryMemoryForm, user=Depends(get_verified_user)
+):
+    query_embedding = request.app.state.EMBEDDING_FUNCTION(form_data.content)
+    collection = CHROMA_CLIENT.get_or_create_collection(name=f"user-memory-{user.id}")
+
+    results = collection.query(
+        query_embeddings=[query_embedding],
+        n_results=1,  # how many results to return
+    )
+
+    return results
+
+
+############################
+# ResetMemoryFromVectorDB
+############################
+@router.get("/reset", response_model=bool)
+async def reset_memory_from_vector_db(
+    request: Request, user=Depends(get_verified_user)
+):
+    CHROMA_CLIENT.delete_collection(f"user-memory-{user.id}")
+    collection = CHROMA_CLIENT.get_or_create_collection(name=f"user-memory-{user.id}")
+
+    memories = Memories.get_memories_by_user_id(user.id)
+    for memory in memories:
+        memory_embedding = request.app.state.EMBEDDING_FUNCTION(memory.content)
+        collection.upsert(
+            documents=[memory.content],
+            ids=[memory.id],
+            embeddings=[memory_embedding],
+        )
+    return True
+
+
+############################
+# DeleteMemoriesByUserId
+############################
+
+
+@router.delete("/user", response_model=bool)
+async def delete_memory_by_user_id(user=Depends(get_verified_user)):
+    result = Memories.delete_memories_by_user_id(user.id)
+
+    if result:
+        try:
+            CHROMA_CLIENT.delete_collection(f"user-memory-{user.id}")
+        except Exception as e:
+            log.error(e)
+        return True
+
+    return False
+
+
+############################
+# DeleteMemoryById
+############################
+
+
+@router.delete("/{memory_id}", response_model=bool)
+async def delete_memory_by_id(memory_id: str, user=Depends(get_verified_user)):
+    result = Memories.delete_memory_by_id_and_user_id(memory_id, user.id)
+
+    if result:
+        collection = CHROMA_CLIENT.get_or_create_collection(
+            name=f"user-memory-{user.id}"
+        )
+        collection.delete(ids=[memory_id])
+        return True
+
+    return False

backend/apps/web/routers/users.py

@@ -11,8 +11,9 @@ import logging
 
 from apps.web.models.users import UserModel, UserUpdateForm, UserRoleUpdateForm, Users
 from apps.web.models.auths import Auths
+from apps.web.models.chats import Chats
 
-from utils.utils import get_current_user, get_password_hash, get_admin_user
+from utils.utils import get_verified_user, get_password_hash, get_admin_user
 from constants import ERROR_MESSAGES
 
 from config import SRC_LOG_LEVELS
@@ -39,15 +40,15 @@ async def get_users(skip: int = 0, limit: int = 50, user=Depends(get_admin_user)
 
 @router.get("/permissions/user")
 async def get_user_permissions(request: Request, user=Depends(get_admin_user)):
-    return request.app.state.USER_PERMISSIONS
+    return request.app.state.config.USER_PERMISSIONS
 
 
 @router.post("/permissions/user")
 async def update_user_permissions(
     request: Request, form_data: dict, user=Depends(get_admin_user)
 ):
-    request.app.state.USER_PERMISSIONS = form_data
-    return request.app.state.USER_PERMISSIONS
+    request.app.state.config.USER_PERMISSIONS = form_data
+    return request.app.state.config.USER_PERMISSIONS
 
 
 ############################
@@ -58,7 +59,7 @@ async def update_user_permissions(
 @router.post("/update/role", response_model=Optional[UserModel])
 async def update_user_role(form_data: UserRoleUpdateForm, user=Depends(get_admin_user)):
 
-    if user.id != form_data.id:
+    if user.id != form_data.id and form_data.id != Users.get_first_user().id:
         return Users.update_user_role_by_id(form_data.id, form_data.role)
 
     raise HTTPException(
@@ -67,6 +68,41 @@ async def update_user_role(form_data: UserRoleUpdateForm, user=Depends(get_admin
     )
 
 
+############################
+# GetUserById
+############################
+
+
+class UserResponse(BaseModel):
+    name: str
+    profile_image_url: str
+
+
+@router.get("/{user_id}", response_model=UserResponse)
+async def get_user_by_id(user_id: str, user=Depends(get_verified_user)):
+
+    if user_id.startswith("shared-"):
+        chat_id = user_id.replace("shared-", "")
+        chat = Chats.get_chat_by_id(chat_id)
+        if chat:
+            user_id = chat.user_id
+        else:
+            raise HTTPException(
+                status_code=status.HTTP_400_BAD_REQUEST,
+                detail=ERROR_MESSAGES.USER_NOT_FOUND,
+            )
+
+    user = Users.get_user_by_id(user_id)
+
+    if user:
+        return UserResponse(name=user.name, profile_image_url=user.profile_image_url)
+    else:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.USER_NOT_FOUND,
+        )
+
+
 ############################
 # UpdateUserById
 ############################

backend/apps/web/routers/utils.py

@@ -1,5 +1,6 @@
 from fastapi import APIRouter, UploadFile, File, Response
 from fastapi import Depends, HTTPException, status
+from peewee import SqliteDatabase
 from starlette.responses import StreamingResponse, FileResponse
 from pydantic import BaseModel
 
@@ -7,11 +8,11 @@ from pydantic import BaseModel
 from fpdf import FPDF
 import markdown
 
-
+from apps.web.internal.db import DB
 from utils.utils import get_admin_user
 from utils.misc import calculate_sha256, get_gravatar_url
 
-from config import OLLAMA_BASE_URLS, DATA_DIR, UPLOAD_DIR
+from config import OLLAMA_BASE_URLS, DATA_DIR, UPLOAD_DIR, ENABLE_ADMIN_EXPORT
 from constants import ERROR_MESSAGES
 from typing import List
 
@@ -91,9 +92,18 @@ async def download_chat_as_pdf(
 
 @router.get("/db/download")
 async def download_db(user=Depends(get_admin_user)):
-
+    if not ENABLE_ADMIN_EXPORT:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
+    if not isinstance(DB, SqliteDatabase):
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DB_NOT_SQLITE,
+        )
     return FileResponse(
-        f"{DATA_DIR}/webui.db",
+        DB.database,
         media_type="application/octet-stream",
         filename="webui.db",
     )

backend/config.py

@@ -5,6 +5,7 @@ import chromadb
 from chromadb import Settings
 from base64 import b64encode
 from bs4 import BeautifulSoup
+from typing import TypeVar, Generic, Union
 
 from pathlib import Path
 import json
@@ -17,17 +18,68 @@ import shutil
 from secrets import token_bytes
 from constants import ERROR_MESSAGES
 
+####################################
+# Load .env file
+####################################
 
 try:
     from dotenv import load_dotenv, find_dotenv
 
     load_dotenv(find_dotenv("../.env"))
 except ImportError:
-    log.warning("dotenv not installed, skipping...")
+    print("dotenv not installed, skipping...")
+
+
+####################################
+# LOGGING
+####################################
+
+log_levels = ["CRITICAL", "ERROR", "WARNING", "INFO", "DEBUG"]
+
+GLOBAL_LOG_LEVEL = os.environ.get("GLOBAL_LOG_LEVEL", "").upper()
+if GLOBAL_LOG_LEVEL in log_levels:
+    logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL, force=True)
+else:
+    GLOBAL_LOG_LEVEL = "INFO"
+
+log = logging.getLogger(__name__)
+log.info(f"GLOBAL_LOG_LEVEL: {GLOBAL_LOG_LEVEL}")
+
+log_sources = [
+    "AUDIO",
+    "COMFYUI",
+    "CONFIG",
+    "DB",
+    "IMAGES",
+    "LITELLM",
+    "MAIN",
+    "MODELS",
+    "OLLAMA",
+    "OPENAI",
+    "RAG",
+    "WEBHOOK",
+]
+
+SRC_LOG_LEVELS = {}
+
+for source in log_sources:
+    log_env_var = source + "_LOG_LEVEL"
+    SRC_LOG_LEVELS[source] = os.environ.get(log_env_var, "").upper()
+    if SRC_LOG_LEVELS[source] not in log_levels:
+        SRC_LOG_LEVELS[source] = GLOBAL_LOG_LEVEL
+    log.info(f"{log_env_var}: {SRC_LOG_LEVELS[source]}")
+
+log.setLevel(SRC_LOG_LEVELS["CONFIG"])
 
 WEBUI_NAME = os.environ.get("WEBUI_NAME", "Open WebUI")
+if WEBUI_NAME != "Open WebUI":
+    WEBUI_NAME += " (Open WebUI)"
+
+WEBUI_URL = os.environ.get("WEBUI_URL", "http://localhost:3000")
+
 WEBUI_FAVICON_URL = "https://openwebui.com/favicon.png"
 
+
 ####################################
 # ENV (dev,test,prod)
 ####################################
@@ -101,6 +153,13 @@ for version in soup.find_all("h2"):
 
 CHANGELOG = changelog_json
 
+
+####################################
+# WEBUI_VERSION
+####################################
+
+WEBUI_VERSION = os.environ.get("WEBUI_VERSION", "v1.0.0-alpha.100")
+
 ####################################
 # DATA/FRONTEND BUILD DIR
 ####################################
@@ -114,54 +173,119 @@ try:
 except:
     CONFIG_DATA = {}
 
+
+####################################
+# Config helpers
+####################################
+
+
+def save_config():
+    try:
+        with open(f"{DATA_DIR}/config.json", "w") as f:
+            json.dump(CONFIG_DATA, f, indent="\t")
+    except Exception as e:
+        log.exception(e)
+
+
+def get_config_value(config_path: str):
+    path_parts = config_path.split(".")
+    cur_config = CONFIG_DATA
+    for key in path_parts:
+        if key in cur_config:
+            cur_config = cur_config[key]
+        else:
+            return None
+    return cur_config
+
+
+T = TypeVar("T")
+
+
+class PersistentConfig(Generic[T]):
+    def __init__(self, env_name: str, config_path: str, env_value: T):
+        self.env_name = env_name
+        self.config_path = config_path
+        self.env_value = env_value
+        self.config_value = get_config_value(config_path)
+        if self.config_value is not None:
+            log.info(f"'{env_name}' loaded from config.json")
+            self.value = self.config_value
+        else:
+            self.value = env_value
+
+    def __str__(self):
+        return str(self.value)
+
+    @property
+    def __dict__(self):
+        raise TypeError(
+            "PersistentConfig object cannot be converted to dict, use config_get or .value instead."
+        )
+
+    def __getattribute__(self, item):
+        if item == "__dict__":
+            raise TypeError(
+                "PersistentConfig object cannot be converted to dict, use config_get or .value instead."
+            )
+        return super().__getattribute__(item)
+
+    def save(self):
+        # Don't save if the value is the same as the env value and the config value
+        if self.env_value == self.value:
+            if self.config_value == self.value:
+                return
+        log.info(f"Saving '{self.env_name}' to config.json")
+        path_parts = self.config_path.split(".")
+        config = CONFIG_DATA
+        for key in path_parts[:-1]:
+            if key not in config:
+                config[key] = {}
+            config = config[key]
+        config[path_parts[-1]] = self.value
+        save_config()
+        self.config_value = self.value
+
+
+class AppConfig:
+    _state: dict[str, PersistentConfig]
+
+    def __init__(self):
+        super().__setattr__("_state", {})
+
+    def __setattr__(self, key, value):
+        if isinstance(value, PersistentConfig):
+            self._state[key] = value
+        else:
+            self._state[key].value = value
+            self._state[key].save()
+
+    def __getattr__(self, key):
+        return self._state[key].value
+
+
+####################################
+# WEBUI_AUTH (Required for security)
+####################################
+
+WEBUI_AUTH = os.environ.get("WEBUI_AUTH", "True").lower() == "true"
+WEBUI_AUTH_TRUSTED_EMAIL_HEADER = os.environ.get(
+    "WEBUI_AUTH_TRUSTED_EMAIL_HEADER", None
+)
+JWT_EXPIRES_IN = PersistentConfig(
+    "JWT_EXPIRES_IN", "auth.jwt_expiry", os.environ.get("JWT_EXPIRES_IN", "-1")
+)
+
 ####################################
 # Static DIR
 ####################################
 
 STATIC_DIR = str(Path(os.getenv("STATIC_DIR", "./static")).resolve())
 
-shutil.copyfile(f"{FRONTEND_BUILD_DIR}/favicon.png", f"{STATIC_DIR}/favicon.png")
-
-####################################
-# LOGGING
-####################################
-log_levels = ["CRITICAL", "ERROR", "WARNING", "INFO", "DEBUG"]
-
-GLOBAL_LOG_LEVEL = os.environ.get("GLOBAL_LOG_LEVEL", "").upper()
-if GLOBAL_LOG_LEVEL in log_levels:
-    logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL, force=True)
+frontend_favicon = f"{FRONTEND_BUILD_DIR}/favicon.png"
+if os.path.exists(frontend_favicon):
+    shutil.copyfile(frontend_favicon, f"{STATIC_DIR}/favicon.png")
 else:
-    GLOBAL_LOG_LEVEL = "INFO"
-
-log = logging.getLogger(__name__)
-log.info(f"GLOBAL_LOG_LEVEL: {GLOBAL_LOG_LEVEL}")
-
-log_sources = [
-    "AUDIO",
-    "COMFYUI",
-    "CONFIG",
-    "DB",
-    "IMAGES",
-    "LITELLM",
-    "MAIN",
-    "MODELS",
-    "OLLAMA",
-    "OPENAI",
-    "RAG",
-    "WEBHOOK",
-]
-
-SRC_LOG_LEVELS = {}
-
-for source in log_sources:
-    log_env_var = source + "_LOG_LEVEL"
-    SRC_LOG_LEVELS[source] = os.environ.get(log_env_var, "").upper()
-    if SRC_LOG_LEVELS[source] not in log_levels:
-        SRC_LOG_LEVELS[source] = GLOBAL_LOG_LEVEL
-    log.info(f"{log_env_var}: {SRC_LOG_LEVELS[source]}")
-
-log.setLevel(SRC_LOG_LEVELS["CONFIG"])
-
+    logging.warning(f"Frontend favicon not found at {frontend_favicon}")
 
 ####################################
 # CUSTOM_NAME
@@ -191,9 +315,6 @@ if CUSTOM_NAME:
     except Exception as e:
         log.exception(e)
         pass
-else:
-    if WEBUI_NAME != "Open WebUI":
-        WEBUI_NAME += " (Open WebUI)"
 
 
 ####################################
@@ -216,7 +337,7 @@ Path(CACHE_DIR).mkdir(parents=True, exist_ok=True)
 # Docs DIR
 ####################################
 
-DOCS_DIR = f"{DATA_DIR}/docs"
+DOCS_DIR = os.getenv("DOCS_DIR", f"{DATA_DIR}/docs")
 Path(DOCS_DIR).mkdir(parents=True, exist_ok=True)
 
 
@@ -288,12 +409,22 @@ OLLAMA_BASE_URLS = os.environ.get("OLLAMA_BASE_URLS", "")
 OLLAMA_BASE_URLS = OLLAMA_BASE_URLS if OLLAMA_BASE_URLS != "" else OLLAMA_BASE_URL
 
 OLLAMA_BASE_URLS = [url.strip() for url in OLLAMA_BASE_URLS.split(";")]
-
+OLLAMA_BASE_URLS = PersistentConfig(
+    "OLLAMA_BASE_URLS", "ollama.base_urls", OLLAMA_BASE_URLS
+)
 
 ####################################
 # OPENAI_API
 ####################################
 
+
+ENABLE_OPENAI_API = PersistentConfig(
+    "ENABLE_OPENAI_API",
+    "openai.enable",
+    os.environ.get("ENABLE_OPENAI_API", "True").lower() == "true",
+)
+
+
 OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "")
 OPENAI_API_BASE_URL = os.environ.get("OPENAI_API_BASE_URL", "")
 
@@ -305,7 +436,9 @@ OPENAI_API_KEYS = os.environ.get("OPENAI_API_KEYS", "")
 OPENAI_API_KEYS = OPENAI_API_KEYS if OPENAI_API_KEYS != "" else OPENAI_API_KEY
 
 OPENAI_API_KEYS = [url.strip() for url in OPENAI_API_KEYS.split(";")]
-
+OPENAI_API_KEYS = PersistentConfig(
+    "OPENAI_API_KEYS", "openai.api_keys", OPENAI_API_KEYS
+)
 
 OPENAI_API_BASE_URLS = os.environ.get("OPENAI_API_BASE_URLS", "")
 OPENAI_API_BASE_URLS = (
@@ -316,21 +449,42 @@ OPENAI_API_BASE_URLS = [
     url.strip() if url != "" else "https://api.openai.com/v1"
     for url in OPENAI_API_BASE_URLS.split(";")
 ]
+OPENAI_API_BASE_URLS = PersistentConfig(
+    "OPENAI_API_BASE_URLS", "openai.api_base_urls", OPENAI_API_BASE_URLS
+)
+
+OPENAI_API_KEY = ""
+
+try:
+    OPENAI_API_KEY = OPENAI_API_KEYS.value[
+        OPENAI_API_BASE_URLS.value.index("https://api.openai.com/v1")
+    ]
+except:
+    pass
+
+OPENAI_API_BASE_URL = "https://api.openai.com/v1"
 
 ####################################
 # WEBUI
 ####################################
 
-ENABLE_SIGNUP = os.environ.get("ENABLE_SIGNUP", "True").lower() == "true"
-DEFAULT_MODELS = os.environ.get("DEFAULT_MODELS", None)
+ENABLE_SIGNUP = PersistentConfig(
+    "ENABLE_SIGNUP",
+    "ui.enable_signup",
+    (
+        False
+        if not WEBUI_AUTH
+        else os.environ.get("ENABLE_SIGNUP", "True").lower() == "true"
+    ),
+)
+DEFAULT_MODELS = PersistentConfig(
+    "DEFAULT_MODELS", "ui.default_models", os.environ.get("DEFAULT_MODELS", None)
+)
 
-
-DEFAULT_PROMPT_SUGGESTIONS = (
-    CONFIG_DATA["ui"]["prompt_suggestions"]
-    if "ui" in CONFIG_DATA
-    and "prompt_suggestions" in CONFIG_DATA["ui"]
-    and type(CONFIG_DATA["ui"]["prompt_suggestions"]) is list
-    else [
+DEFAULT_PROMPT_SUGGESTIONS = PersistentConfig(
+    "DEFAULT_PROMPT_SUGGESTIONS",
+    "ui.prompt_suggestions",
+    [
         {
             "title": ["Help me study", "vocabulary for a college entrance exam"],
             "content": "Help me study vocabulary: write a sentence for me to fill in the blank, and I'll try to pick the correct option.",
@@ -347,40 +501,54 @@ DEFAULT_PROMPT_SUGGESTIONS = (
             "title": ["Show me a code snippet", "of a website's sticky header"],
             "content": "Show me a code snippet of a website's sticky header in CSS and JavaScript.",
         },
-    ]
+        {
+            "title": [
+                "Explain options trading",
+                "if I'm familiar with buying and selling stocks",
+            ],
+            "content": "Explain options trading in simple terms if I'm familiar with buying and selling stocks.",
+        },
+        {
+            "title": ["Overcome procrastination", "give me tips"],
+            "content": "Could you start by asking me about instances when I procrastinate the most and then give me some suggestions to overcome it?",
+        },
+    ],
 )
 
-
-DEFAULT_USER_ROLE = os.getenv("DEFAULT_USER_ROLE", "pending")
+DEFAULT_USER_ROLE = PersistentConfig(
+    "DEFAULT_USER_ROLE",
+    "ui.default_user_role",
+    os.getenv("DEFAULT_USER_ROLE", "pending"),
+)
 
 USER_PERMISSIONS_CHAT_DELETION = (
     os.environ.get("USER_PERMISSIONS_CHAT_DELETION", "True").lower() == "true"
 )
 
-USER_PERMISSIONS = {"chat": {"deletion": USER_PERMISSIONS_CHAT_DELETION}}
-
-
-MODEL_FILTER_ENABLED = os.environ.get("MODEL_FILTER_ENABLED", "False").lower() == "true"
-MODEL_FILTER_LIST = os.environ.get("MODEL_FILTER_LIST", "")
-MODEL_FILTER_LIST = [model.strip() for model in MODEL_FILTER_LIST.split(";")]
-
-WEBHOOK_URL = os.environ.get("WEBHOOK_URL", "")
-
-####################################
-# WEBUI_VERSION
-####################################
-
-WEBUI_VERSION = os.environ.get("WEBUI_VERSION", "v1.0.0-alpha.100")
-
-####################################
-# WEBUI_AUTH (Required for security)
-####################################
-
-WEBUI_AUTH = True
-WEBUI_AUTH_TRUSTED_EMAIL_HEADER = os.environ.get(
-    "WEBUI_AUTH_TRUSTED_EMAIL_HEADER", None
+USER_PERMISSIONS = PersistentConfig(
+    "USER_PERMISSIONS",
+    "ui.user_permissions",
+    {"chat": {"deletion": USER_PERMISSIONS_CHAT_DELETION}},
 )
 
+ENABLE_MODEL_FILTER = PersistentConfig(
+    "ENABLE_MODEL_FILTER",
+    "model_filter.enable",
+    os.environ.get("ENABLE_MODEL_FILTER", "False").lower() == "true",
+)
+MODEL_FILTER_LIST = os.environ.get("MODEL_FILTER_LIST", "")
+MODEL_FILTER_LIST = PersistentConfig(
+    "MODEL_FILTER_LIST",
+    "model_filter.list",
+    [model.strip() for model in MODEL_FILTER_LIST.split(";")],
+)
+
+WEBHOOK_URL = PersistentConfig(
+    "WEBHOOK_URL", "webhook_url", os.environ.get("WEBHOOK_URL", "")
+)
+
+ENABLE_ADMIN_EXPORT = os.environ.get("ENABLE_ADMIN_EXPORT", "True").lower() == "true"
+
 ####################################
 # WEBUI_SECRET_KEY
 ####################################
@@ -400,16 +568,106 @@ if WEBUI_AUTH and WEBUI_SECRET_KEY == "":
 ####################################
 
 CHROMA_DATA_PATH = f"{DATA_DIR}/vector_db"
-# this uses the model defined in the Dockerfile ENV variable. If you dont use docker or docker based deployments such as k8s, the default embedding model will be used (all-MiniLM-L6-v2)
-RAG_EMBEDDING_MODEL = os.environ.get("RAG_EMBEDDING_MODEL", "all-MiniLM-L6-v2")
-log.info(f"Embedding model set: {RAG_EMBEDDING_MODEL}"),
+CHROMA_TENANT = os.environ.get("CHROMA_TENANT", chromadb.DEFAULT_TENANT)
+CHROMA_DATABASE = os.environ.get("CHROMA_DATABASE", chromadb.DEFAULT_DATABASE)
+CHROMA_HTTP_HOST = os.environ.get("CHROMA_HTTP_HOST", "")
+CHROMA_HTTP_PORT = int(os.environ.get("CHROMA_HTTP_PORT", "8000"))
+# Comma-separated list of header=value pairs
+CHROMA_HTTP_HEADERS = os.environ.get("CHROMA_HTTP_HEADERS", "")
+if CHROMA_HTTP_HEADERS:
+    CHROMA_HTTP_HEADERS = dict(
+        [pair.split("=") for pair in CHROMA_HTTP_HEADERS.split(",")]
+    )
+else:
+    CHROMA_HTTP_HEADERS = None
+CHROMA_HTTP_SSL = os.environ.get("CHROMA_HTTP_SSL", "false").lower() == "true"
+# this uses the model defined in the Dockerfile ENV variable. If you dont use docker or docker based deployments such as k8s, the default embedding model will be used (sentence-transformers/all-MiniLM-L6-v2)
+
+RAG_TOP_K = PersistentConfig(
+    "RAG_TOP_K", "rag.top_k", int(os.environ.get("RAG_TOP_K", "5"))
+)
+RAG_RELEVANCE_THRESHOLD = PersistentConfig(
+    "RAG_RELEVANCE_THRESHOLD",
+    "rag.relevance_threshold",
+    float(os.environ.get("RAG_RELEVANCE_THRESHOLD", "0.0")),
+)
+
+ENABLE_RAG_HYBRID_SEARCH = PersistentConfig(
+    "ENABLE_RAG_HYBRID_SEARCH",
+    "rag.enable_hybrid_search",
+    os.environ.get("ENABLE_RAG_HYBRID_SEARCH", "").lower() == "true",
+)
+
+ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = PersistentConfig(
+    "ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION",
+    "rag.enable_web_loader_ssl_verification",
+    os.environ.get("ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION", "True").lower() == "true",
+)
+
+RAG_EMBEDDING_ENGINE = PersistentConfig(
+    "RAG_EMBEDDING_ENGINE",
+    "rag.embedding_engine",
+    os.environ.get("RAG_EMBEDDING_ENGINE", ""),
+)
+
+PDF_EXTRACT_IMAGES = PersistentConfig(
+    "PDF_EXTRACT_IMAGES",
+    "rag.pdf_extract_images",
+    os.environ.get("PDF_EXTRACT_IMAGES", "False").lower() == "true",
+)
+
+RAG_EMBEDDING_MODEL = PersistentConfig(
+    "RAG_EMBEDDING_MODEL",
+    "rag.embedding_model",
+    os.environ.get("RAG_EMBEDDING_MODEL", "sentence-transformers/all-MiniLM-L6-v2"),
+)
+log.info(f"Embedding model set: {RAG_EMBEDDING_MODEL.value}"),
 
 RAG_EMBEDDING_MODEL_AUTO_UPDATE = (
     os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "").lower() == "true"
 )
 
+RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE = (
+    os.environ.get("RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE", "").lower() == "true"
+)
 
-# device type ebbeding models - "cpu" (default), "cuda" (nvidia gpu required) or "mps" (apple silicon) - choosing this right can lead to better performance
+RAG_RERANKING_MODEL = PersistentConfig(
+    "RAG_RERANKING_MODEL",
+    "rag.reranking_model",
+    os.environ.get("RAG_RERANKING_MODEL", ""),
+)
+if RAG_RERANKING_MODEL.value != "":
+    log.info(f"Reranking model set: {RAG_RERANKING_MODEL.value}"),
+
+RAG_RERANKING_MODEL_AUTO_UPDATE = (
+    os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "").lower() == "true"
+)
+
+RAG_RERANKING_MODEL_TRUST_REMOTE_CODE = (
+    os.environ.get("RAG_RERANKING_MODEL_TRUST_REMOTE_CODE", "").lower() == "true"
+)
+
+
+if CHROMA_HTTP_HOST != "":
+    CHROMA_CLIENT = chromadb.HttpClient(
+        host=CHROMA_HTTP_HOST,
+        port=CHROMA_HTTP_PORT,
+        headers=CHROMA_HTTP_HEADERS,
+        ssl=CHROMA_HTTP_SSL,
+        tenant=CHROMA_TENANT,
+        database=CHROMA_DATABASE,
+        settings=Settings(allow_reset=True, anonymized_telemetry=False),
+    )
+else:
+    CHROMA_CLIENT = chromadb.PersistentClient(
+        path=CHROMA_DATA_PATH,
+        settings=Settings(allow_reset=True, anonymized_telemetry=False),
+        tenant=CHROMA_TENANT,
+        database=CHROMA_DATABASE,
+    )
+
+
+# device type embedding models - "cpu" (default), "cuda" (nvidia gpu required) or "mps" (apple silicon) - choosing this right can lead to better performance
 USE_CUDA = os.environ.get("USE_CUDA_DOCKER", "false")
 
 if USE_CUDA.lower() == "true":
@@ -417,16 +675,16 @@ if USE_CUDA.lower() == "true":
 else:
     DEVICE_TYPE = "cpu"
 
-
-CHROMA_CLIENT = chromadb.PersistentClient(
-    path=CHROMA_DATA_PATH,
-    settings=Settings(allow_reset=True, anonymized_telemetry=False),
+CHUNK_SIZE = PersistentConfig(
+    "CHUNK_SIZE", "rag.chunk_size", int(os.environ.get("CHUNK_SIZE", "1500"))
+)
+CHUNK_OVERLAP = PersistentConfig(
+    "CHUNK_OVERLAP",
+    "rag.chunk_overlap",
+    int(os.environ.get("CHUNK_OVERLAP", "100")),
 )
-CHUNK_SIZE = 1500
-CHUNK_OVERLAP = 100
 
-
-RAG_TEMPLATE = """Use the following context as your learned knowledge, inside <context></context> XML tags.
+DEFAULT_RAG_TEMPLATE = """Use the following context as your learned knowledge, inside <context></context> XML tags.
 <context>
     [context]
 </context>
@@ -436,21 +694,140 @@ When answer to user:
 - If you don't know when you are not sure, ask for clarification.
 Avoid mentioning that you obtained the information from the context.
 And answer according to the language of the user's question.
-        
+
 Given the context information, answer the query.
 Query: [query]"""
 
+RAG_TEMPLATE = PersistentConfig(
+    "RAG_TEMPLATE",
+    "rag.template",
+    os.environ.get("RAG_TEMPLATE", DEFAULT_RAG_TEMPLATE),
+)
+
+RAG_OPENAI_API_BASE_URL = PersistentConfig(
+    "RAG_OPENAI_API_BASE_URL",
+    "rag.openai_api_base_url",
+    os.getenv("RAG_OPENAI_API_BASE_URL", OPENAI_API_BASE_URL),
+)
+RAG_OPENAI_API_KEY = PersistentConfig(
+    "RAG_OPENAI_API_KEY",
+    "rag.openai_api_key",
+    os.getenv("RAG_OPENAI_API_KEY", OPENAI_API_KEY),
+)
+
+ENABLE_RAG_LOCAL_WEB_FETCH = (
+    os.getenv("ENABLE_RAG_LOCAL_WEB_FETCH", "False").lower() == "true"
+)
+
+YOUTUBE_LOADER_LANGUAGE = PersistentConfig(
+    "YOUTUBE_LOADER_LANGUAGE",
+    "rag.youtube_loader_language",
+    os.getenv("YOUTUBE_LOADER_LANGUAGE", "en").split(","),
+)
+
 ####################################
 # Transcribe
 ####################################
 
 WHISPER_MODEL = os.getenv("WHISPER_MODEL", "base")
 WHISPER_MODEL_DIR = os.getenv("WHISPER_MODEL_DIR", f"{CACHE_DIR}/whisper/models")
+WHISPER_MODEL_AUTO_UPDATE = (
+    os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true"
+)
 
 
 ####################################
 # Images
 ####################################
 
-AUTOMATIC1111_BASE_URL = os.getenv("AUTOMATIC1111_BASE_URL", "")
-COMFYUI_BASE_URL = os.getenv("COMFYUI_BASE_URL", "")
+IMAGE_GENERATION_ENGINE = PersistentConfig(
+    "IMAGE_GENERATION_ENGINE",
+    "image_generation.engine",
+    os.getenv("IMAGE_GENERATION_ENGINE", ""),
+)
+
+ENABLE_IMAGE_GENERATION = PersistentConfig(
+    "ENABLE_IMAGE_GENERATION",
+    "image_generation.enable",
+    os.environ.get("ENABLE_IMAGE_GENERATION", "").lower() == "true",
+)
+AUTOMATIC1111_BASE_URL = PersistentConfig(
+    "AUTOMATIC1111_BASE_URL",
+    "image_generation.automatic1111.base_url",
+    os.getenv("AUTOMATIC1111_BASE_URL", ""),
+)
+
+COMFYUI_BASE_URL = PersistentConfig(
+    "COMFYUI_BASE_URL",
+    "image_generation.comfyui.base_url",
+    os.getenv("COMFYUI_BASE_URL", ""),
+)
+
+IMAGES_OPENAI_API_BASE_URL = PersistentConfig(
+    "IMAGES_OPENAI_API_BASE_URL",
+    "image_generation.openai.api_base_url",
+    os.getenv("IMAGES_OPENAI_API_BASE_URL", OPENAI_API_BASE_URL),
+)
+IMAGES_OPENAI_API_KEY = PersistentConfig(
+    "IMAGES_OPENAI_API_KEY",
+    "image_generation.openai.api_key",
+    os.getenv("IMAGES_OPENAI_API_KEY", OPENAI_API_KEY),
+)
+
+IMAGE_SIZE = PersistentConfig(
+    "IMAGE_SIZE", "image_generation.size", os.getenv("IMAGE_SIZE", "512x512")
+)
+
+IMAGE_STEPS = PersistentConfig(
+    "IMAGE_STEPS", "image_generation.steps", int(os.getenv("IMAGE_STEPS", 50))
+)
+
+IMAGE_GENERATION_MODEL = PersistentConfig(
+    "IMAGE_GENERATION_MODEL",
+    "image_generation.model",
+    os.getenv("IMAGE_GENERATION_MODEL", ""),
+)
+
+####################################
+# Audio
+####################################
+
+AUDIO_OPENAI_API_BASE_URL = PersistentConfig(
+    "AUDIO_OPENAI_API_BASE_URL",
+    "audio.openai.api_base_url",
+    os.getenv("AUDIO_OPENAI_API_BASE_URL", OPENAI_API_BASE_URL),
+)
+AUDIO_OPENAI_API_KEY = PersistentConfig(
+    "AUDIO_OPENAI_API_KEY",
+    "audio.openai.api_key",
+    os.getenv("AUDIO_OPENAI_API_KEY", OPENAI_API_KEY),
+)
+AUDIO_OPENAI_API_MODEL = PersistentConfig(
+    "AUDIO_OPENAI_API_MODEL",
+    "audio.openai.api_model",
+    os.getenv("AUDIO_OPENAI_API_MODEL", "tts-1"),
+)
+AUDIO_OPENAI_API_VOICE = PersistentConfig(
+    "AUDIO_OPENAI_API_VOICE",
+    "audio.openai.api_voice",
+    os.getenv("AUDIO_OPENAI_API_VOICE", "alloy"),
+)
+
+####################################
+# LiteLLM
+####################################
+
+
+ENABLE_LITELLM = os.environ.get("ENABLE_LITELLM", "True").lower() == "true"
+
+LITELLM_PROXY_PORT = int(os.getenv("LITELLM_PROXY_PORT", "14365"))
+if LITELLM_PROXY_PORT < 0 or LITELLM_PROXY_PORT > 65535:
+    raise ValueError("Invalid port number for LITELLM_PROXY_PORT")
+LITELLM_PROXY_HOST = os.getenv("LITELLM_PROXY_HOST", "127.0.0.1")
+
+
+####################################
+# Database
+####################################
+
+DATABASE_URL = os.environ.get("DATABASE_URL", f"sqlite:///{DATA_DIR}/webui.db")

backend/constants.py

@@ -3,6 +3,10 @@ from enum import Enum
 
 class MESSAGES(str, Enum):
     DEFAULT = lambda msg="": f"{msg if msg else ''}"
+    MODEL_ADDED = lambda model="": f"The model '{model}' has been added successfully."
+    MODEL_DELETED = (
+        lambda model="": f"The model '{model}' has been deleted successfully."
+    )
 
 
 class WEBHOOK_MESSAGES(str, Enum):
@@ -38,6 +42,9 @@ class ERROR_MESSAGES(str, Enum):
         "The password provided is incorrect. Please check for typos and try again."
     )
     INVALID_TRUSTED_HEADER = "Your provider has not provided a trusted header. Please contact your administrator for assistance."
+
+    EXISTING_USERS = "You can't turn off authentication because there are existing users. If you want to disable WEBUI_AUTH, make sure your web interface doesn't have any existing users and is a fresh installation."
+
     UNAUTHORIZED = "401 Unauthorized"
     ACCESS_PROHIBITED = "You do not have permission to access this resource. Please contact your administrator for assistance."
     ACTION_PROHIBITED = (
@@ -65,3 +72,9 @@ class ERROR_MESSAGES(str, Enum):
     CREATE_API_KEY_ERROR = "Oops! Something went wrong while creating your API key. Please try again later. If the issue persists, contact support for assistance."
 
     EMPTY_CONTENT = "The content provided is empty. Please ensure that there is text or data present before proceeding."
+
+    DB_NOT_SQLITE = "This feature is only available when running with SQLite databases."
+
+    INVALID_URL = (
+        "Oops! The URL you provided is invalid. Please double-check and try again."
+    )

backend/data/config.json

@@ -18,6 +18,18 @@
 			{
 				"title": ["Show me a code snippet", "of a website's sticky header"],
 				"content": "Show me a code snippet of a website's sticky header in CSS and JavaScript."
+			},
+			{
+				"title": ["Explain options trading", "if I'm familiar with buying and selling stocks"],
+				"content": "Explain options trading in simple terms if I'm familiar with buying and selling stocks."
+			},
+			{
+				"title": ["Overcome procrastination", "give me tips"],
+				"content": "Could you start by asking me about instances when I procrastinate the most and then give me some suggestions to overcome it?"
+			},
+			{
+				"title": ["Grammar check", "rewrite it for better readability "],
+				"content": "Check the following sentence for grammar and clarity: \"[sentence]\". Rewrite it for better readability while maintaining its original meaning."
 			}
 		]
 	}

backend/main.py

@@ -1,3 +1,4 @@
+from contextlib import asynccontextmanager
 from bs4 import BeautifulSoup
 import json
 import markdown
@@ -15,17 +16,24 @@ from fastapi.middleware.wsgi import WSGIMiddleware
 from fastapi.middleware.cors import CORSMiddleware
 from starlette.exceptions import HTTPException as StarletteHTTPException
 from starlette.middleware.base import BaseHTTPMiddleware
-
+from starlette.responses import StreamingResponse, Response
 
 from apps.ollama.main import app as ollama_app
 from apps.openai.main import app as openai_app
 
-from apps.litellm.main import app as litellm_app, startup as litellm_app_startup
+from apps.litellm.main import (
+    app as litellm_app,
+    start_litellm_background,
+    shutdown_litellm_background,
+)
+
+
 from apps.audio.main import app as audio_app
 from apps.images.main import app as images_app
 from apps.rag.main import app as rag_app
 from apps.web.main import app as webui_app
 
+import asyncio
 from pydantic import BaseModel
 from typing import List
 
@@ -36,17 +44,22 @@ from apps.rag.utils import rag_messages
 from config import (
     CONFIG_DATA,
     WEBUI_NAME,
+    WEBUI_URL,
+    WEBUI_AUTH,
     ENV,
     VERSION,
     CHANGELOG,
     FRONTEND_BUILD_DIR,
     CACHE_DIR,
     STATIC_DIR,
-    MODEL_FILTER_ENABLED,
+    ENABLE_LITELLM,
+    ENABLE_MODEL_FILTER,
     MODEL_FILTER_LIST,
     GLOBAL_LOG_LEVEL,
     SRC_LOG_LEVELS,
     WEBHOOK_URL,
+    ENABLE_ADMIN_EXPORT,
+    AppConfig,
 )
 from constants import ERROR_MESSAGES
 
@@ -67,7 +80,7 @@ class SPAStaticFiles(StaticFiles):
 
 
 print(
-    f"""
+    rf"""
   ___                    __        __   _     _   _ ___ 
  / _ \ _ __   ___ _ __   \ \      / /__| |__ | | | |_ _|
 | | | | '_ \ / _ \ '_ \   \ \ /\ / / _ \ '_ \| | | || | 
@@ -81,18 +94,45 @@ https://github.com/open-webui/open-webui
 """
 )
 
-app = FastAPI(docs_url="/docs" if ENV == "dev" else None, redoc_url=None)
 
-app.state.MODEL_FILTER_ENABLED = MODEL_FILTER_ENABLED
-app.state.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    if ENABLE_LITELLM:
+        asyncio.create_task(start_litellm_background())
+    yield
+    if ENABLE_LITELLM:
+        await shutdown_litellm_background()
 
-app.state.WEBHOOK_URL = WEBHOOK_URL
+
+app = FastAPI(
+    docs_url="/docs" if ENV == "dev" else None, redoc_url=None, lifespan=lifespan
+)
+
+app.state.config = AppConfig()
+app.state.config.ENABLE_MODEL_FILTER = ENABLE_MODEL_FILTER
+app.state.config.MODEL_FILTER_LIST = MODEL_FILTER_LIST
+
+app.state.config.WEBHOOK_URL = WEBHOOK_URL
 
 origins = ["*"]
 
 
+# Custom middleware to add security headers
+# class SecurityHeadersMiddleware(BaseHTTPMiddleware):
+#     async def dispatch(self, request: Request, call_next):
+#         response: Response = await call_next(request)
+#         response.headers["Cross-Origin-Opener-Policy"] = "same-origin"
+#         response.headers["Cross-Origin-Embedder-Policy"] = "require-corp"
+#         return response
+
+
+# app.add_middleware(SecurityHeadersMiddleware)
+
+
 class RAGMiddleware(BaseHTTPMiddleware):
     async def dispatch(self, request: Request, call_next):
+        return_citations = False
+
         if request.method == "POST" and (
             "/api/chat" in request.url.path or "/chat/completions" in request.url.path
         ):
@@ -105,20 +145,29 @@ class RAGMiddleware(BaseHTTPMiddleware):
             # Parse string to JSON
             data = json.loads(body_str) if body_str else {}
 
+            return_citations = data.get("citations", False)
+            if "citations" in data:
+                del data["citations"]
+
             # Example: Add a new key-value pair or modify existing ones
             # data["modified"] = True  # Example modification
             if "docs" in data:
                 data = {**data}
-                data["messages"] = rag_messages(
-                    data["docs"],
-                    data["messages"],
-                    rag_app.state.RAG_TEMPLATE,
-                    rag_app.state.TOP_K,
-                    rag_app.state.sentence_transformer_ef,
+                data["messages"], citations = rag_messages(
+                    docs=data["docs"],
+                    messages=data["messages"],
+                    template=rag_app.state.config.RAG_TEMPLATE,
+                    embedding_function=rag_app.state.EMBEDDING_FUNCTION,
+                    k=rag_app.state.config.TOP_K,
+                    reranking_function=rag_app.state.sentence_transformer_rf,
+                    r=rag_app.state.config.RELEVANCE_THRESHOLD,
+                    hybrid_search=rag_app.state.config.ENABLE_RAG_HYBRID_SEARCH,
                 )
                 del data["docs"]
 
-                log.debug(f"data['messages']: {data['messages']}")
+                log.debug(
+                    f"data['messages']: {data['messages']}, citations: {citations}"
+                )
 
             modified_body_bytes = json.dumps(data).encode("utf-8")
 
@@ -136,11 +185,36 @@ class RAGMiddleware(BaseHTTPMiddleware):
             ]
 
         response = await call_next(request)
+
+        if return_citations:
+            # Inject the citations into the response
+            if isinstance(response, StreamingResponse):
+                # If it's a streaming response, inject it as SSE event or NDJSON line
+                content_type = response.headers.get("Content-Type")
+                if "text/event-stream" in content_type:
+                    return StreamingResponse(
+                        self.openai_stream_wrapper(response.body_iterator, citations),
+                    )
+                if "application/x-ndjson" in content_type:
+                    return StreamingResponse(
+                        self.ollama_stream_wrapper(response.body_iterator, citations),
+                    )
+
         return response
 
     async def _receive(self, body: bytes):
         return {"type": "http.request", "body": body, "more_body": False}
 
+    async def openai_stream_wrapper(self, original_generator, citations):
+        yield f"data: {json.dumps({'citations': citations})}\n\n"
+        async for data in original_generator:
+            yield data
+
+    async def ollama_stream_wrapper(self, original_generator, citations):
+        yield f"{json.dumps({'citations': citations})}\n"
+        async for data in original_generator:
+            yield data
+
 
 app.add_middleware(RAGMiddleware)
 
@@ -164,14 +238,15 @@ async def check_url(request: Request, call_next):
     return response
 
 
-@app.on_event("startup")
-async def on_startup():
-    await litellm_app_startup()
+@app.middleware("http")
+async def update_embedding_function(request: Request, call_next):
+    response = await call_next(request)
+    if "/embedding/update" in request.url.path:
+        webui_app.state.EMBEDDING_FUNCTION = rag_app.state.EMBEDDING_FUNCTION
+    return response
 
 
-app.mount("/api/v1", webui_app)
 app.mount("/litellm/api", litellm_app)
-
 app.mount("/ollama", ollama_app)
 app.mount("/openai/api", openai_app)
 
@@ -179,6 +254,10 @@ app.mount("/images/api/v1", images_app)
 app.mount("/audio/api/v1", audio_app)
 app.mount("/rag/api/v1", rag_app)
 
+app.mount("/api/v1", webui_app)
+
+webui_app.state.EMBEDDING_FUNCTION = rag_app.state.EMBEDDING_FUNCTION
+
 
 @app.get("/api/config")
 async def get_app_config():
@@ -193,19 +272,21 @@ async def get_app_config():
         "status": True,
         "name": WEBUI_NAME,
         "version": VERSION,
+        "auth": WEBUI_AUTH,
         "default_locale": default_locale,
-        "images": images_app.state.ENABLED,
-        "default_models": webui_app.state.DEFAULT_MODELS,
-        "default_prompt_suggestions": webui_app.state.DEFAULT_PROMPT_SUGGESTIONS,
+        "images": images_app.state.config.ENABLED,
+        "default_models": webui_app.state.config.DEFAULT_MODELS,
+        "default_prompt_suggestions": webui_app.state.config.DEFAULT_PROMPT_SUGGESTIONS,
         "trusted_header_auth": bool(webui_app.state.AUTH_TRUSTED_EMAIL_HEADER),
+        "admin_export_enabled": ENABLE_ADMIN_EXPORT,
     }
 
 
 @app.get("/api/config/model/filter")
 async def get_model_filter_config(user=Depends(get_admin_user)):
     return {
-        "enabled": app.state.MODEL_FILTER_ENABLED,
-        "models": app.state.MODEL_FILTER_LIST,
+        "enabled": app.state.config.ENABLE_MODEL_FILTER,
+        "models": app.state.config.MODEL_FILTER_LIST,
     }
 
 
@@ -218,28 +299,28 @@ class ModelFilterConfigForm(BaseModel):
 async def update_model_filter_config(
     form_data: ModelFilterConfigForm, user=Depends(get_admin_user)
 ):
-    app.state.MODEL_FILTER_ENABLED = form_data.enabled
-    app.state.MODEL_FILTER_LIST = form_data.models
+    app.state.config.ENABLE_MODEL_FILTER = form_data.enabled
+    app.state.config.MODEL_FILTER_LIST = form_data.models
 
-    ollama_app.state.MODEL_FILTER_ENABLED = app.state.MODEL_FILTER_ENABLED
-    ollama_app.state.MODEL_FILTER_LIST = app.state.MODEL_FILTER_LIST
+    ollama_app.state.config.ENABLE_MODEL_FILTER = app.state.config.ENABLE_MODEL_FILTER
+    ollama_app.state.config.MODEL_FILTER_LIST = app.state.config.MODEL_FILTER_LIST
 
-    openai_app.state.MODEL_FILTER_ENABLED = app.state.MODEL_FILTER_ENABLED
-    openai_app.state.MODEL_FILTER_LIST = app.state.MODEL_FILTER_LIST
+    openai_app.state.config.ENABLE_MODEL_FILTER = app.state.config.ENABLE_MODEL_FILTER
+    openai_app.state.config.MODEL_FILTER_LIST = app.state.config.MODEL_FILTER_LIST
 
-    litellm_app.state.MODEL_FILTER_ENABLED = app.state.MODEL_FILTER_ENABLED
-    litellm_app.state.MODEL_FILTER_LIST = app.state.MODEL_FILTER_LIST
+    litellm_app.state.ENABLE_MODEL_FILTER = app.state.config.ENABLE_MODEL_FILTER
+    litellm_app.state.MODEL_FILTER_LIST = app.state.config.MODEL_FILTER_LIST
 
     return {
-        "enabled": app.state.MODEL_FILTER_ENABLED,
-        "models": app.state.MODEL_FILTER_LIST,
+        "enabled": app.state.config.ENABLE_MODEL_FILTER,
+        "models": app.state.config.MODEL_FILTER_LIST,
     }
 
 
 @app.get("/api/webhook")
 async def get_webhook_url(user=Depends(get_admin_user)):
     return {
-        "url": app.state.WEBHOOK_URL,
+        "url": app.state.config.WEBHOOK_URL,
     }
 
 
@@ -249,12 +330,12 @@ class UrlForm(BaseModel):
 
 @app.post("/api/webhook")
 async def update_webhook_url(form_data: UrlForm, user=Depends(get_admin_user)):
-    app.state.WEBHOOK_URL = form_data.url
+    app.state.config.WEBHOOK_URL = form_data.url
 
-    webui_app.state.WEBHOOK_URL = app.state.WEBHOOK_URL
+    webui_app.state.WEBHOOK_URL = app.state.config.WEBHOOK_URL
 
     return {
-        "url": app.state.WEBHOOK_URL,
+        "url": app.state.config.WEBHOOK_URL,
     }
 
 
@@ -299,15 +380,40 @@ async def get_manifest_json():
         "background_color": "#343541",
         "theme_color": "#343541",
         "orientation": "portrait-primary",
-        "icons": [{"src": "/favicon.png", "type": "image/png", "sizes": "844x884"}],
+        "icons": [{"src": "/static/logo.png", "type": "image/png", "sizes": "500x500"}],
     }
 
 
+@app.get("/opensearch.xml")
+async def get_opensearch_xml():
+    xml_content = rf"""
+    <OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/" xmlns:moz="http://www.mozilla.org/2006/browser/search/">
+    <ShortName>{WEBUI_NAME}</ShortName>
+    <Description>Search {WEBUI_NAME}</Description>
+    <InputEncoding>UTF-8</InputEncoding>
+    <Image width="16" height="16" type="image/x-icon">{WEBUI_URL}/favicon.png</Image>
+    <Url type="text/html" method="get" template="{WEBUI_URL}/?q={"{searchTerms}"}"/>
+    <moz:SearchForm>{WEBUI_URL}</moz:SearchForm>
+    </OpenSearchDescription>
+    """
+    return Response(content=xml_content, media_type="application/xml")
+
+
+@app.get("/health")
+async def healthcheck():
+    return {"status": True}
+
+
 app.mount("/static", StaticFiles(directory=STATIC_DIR), name="static")
 app.mount("/cache", StaticFiles(directory=CACHE_DIR), name="cache")
 
-app.mount(
-    "/",
-    SPAStaticFiles(directory=FRONTEND_BUILD_DIR, html=True),
-    name="spa-static-files",
-)
+if os.path.exists(FRONTEND_BUILD_DIR):
+    app.mount(
+        "/",
+        SPAStaticFiles(directory=FRONTEND_BUILD_DIR, html=True),
+        name="spa-static-files",
+    )
+else:
+    log.warning(
+        f"Frontend build directory not found at '{FRONTEND_BUILD_DIR}'. Serving API only."
+    )

backend/requirements.txt

@@ -1,53 +1,60 @@
-fastapi
-uvicorn[standard]
-pydantic
-python-multipart
+fastapi==0.109.2
+uvicorn[standard]==0.22.0
+pydantic==2.7.1
+python-multipart==0.0.9
 
-flask
-flask_cors
+Flask==3.0.3
+Flask-Cors==4.0.0
 
-python-socketio
-python-jose
-passlib[bcrypt]
-uuid
+python-socketio==5.11.2
+python-jose==3.3.0
+passlib[bcrypt]==1.7.4
 
-requests
-aiohttp
-peewee
-peewee-migrate
-bcrypt
+requests==2.31.0
+aiohttp==3.9.5
+peewee==3.17.3
+peewee-migrate==1.12.2
+psycopg2-binary==2.9.9
+PyMySQL==1.1.0
+bcrypt==4.1.2
 
-litellm==1.30.7
-boto3
+litellm[proxy]==1.35.28
 
-argon2-cffi
-apscheduler
-google-generativeai
+boto3==1.34.95
 
-langchain
-langchain-community
-fake_useragent
-chromadb
-sentence_transformers
-pypdf
-docx2txt
-unstructured
-markdown
-pypandoc
-pandas
-openpyxl
-pyxlsb
-xlrd
+argon2-cffi==23.1.0
+APScheduler==3.10.4
+google-generativeai==0.5.2
 
-opencv-python-headless
-rapidocr-onnxruntime
+langchain==0.1.16
+langchain-community==0.0.34
+langchain-chroma==0.1.0
 
-fpdf2
+fake-useragent==1.5.1
+chromadb==0.4.24
+sentence-transformers==2.7.0
+pypdf==4.2.0
+docx2txt==0.8
+unstructured==0.11.8
+Markdown==3.6
+pypandoc==1.13
+pandas==2.2.2
+openpyxl==3.1.2
+pyxlsb==1.0.10
+xlrd==2.0.1
+validators==0.28.1
 
-faster-whisper
+opencv-python-headless==4.9.0.80
+rapidocr-onnxruntime==1.2.3
 
-PyJWT
-pyjwt[crypto]
+fpdf2==2.7.8
+rank-bm25==0.2.2
 
-black
-langfuse
+faster-whisper==1.0.1
+
+PyJWT[crypto]==2.8.0
+
+black==24.4.2
+langfuse==2.27.3
+youtube-transcript-api==0.6.2
+pytube

backend/start.sh

@@ -6,8 +6,9 @@ cd "$SCRIPT_DIR" || exit
 KEY_FILE=.webui_secret_key
 
 PORT="${PORT:-8080}"
+HOST="${HOST:-0.0.0.0}"
 if test "$WEBUI_SECRET_KEY $WEBUI_JWT_SECRET_KEY" = " "; then
-  echo "No WEBUI_SECRET_KEY provided"
+  echo "Loading WEBUI_SECRET_KEY from file, not provided as an environment variable."
 
   if ! [ -e "$KEY_FILE" ]; then
     echo "Generating WEBUI_SECRET_KEY"
@@ -29,4 +30,4 @@ if [ "$USE_CUDA_DOCKER" = "true" ]; then
   export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/lib/python3.11/site-packages/torch/lib:/usr/local/lib/python3.11/site-packages/nvidia/cudnn/lib"
 fi
 
-WEBUI_SECRET_KEY="$WEBUI_SECRET_KEY" exec uvicorn main:app --host 0.0.0.0 --port "$PORT" --forwarded-allow-ips '*'
+WEBUI_SECRET_KEY="$WEBUI_SECRET_KEY" exec uvicorn main:app --host "$HOST" --port "$PORT" --forwarded-allow-ips '*'

backend/start_windows.bat

@@ -7,13 +7,13 @@ SET "SCRIPT_DIR=%~dp0"
 cd /d "%SCRIPT_DIR%" || exit /b
 
 SET "KEY_FILE=.webui_secret_key"
-SET "PORT=%PORT:8080%"
+IF "%PORT%"=="" SET PORT=8080
 SET "WEBUI_SECRET_KEY=%WEBUI_SECRET_KEY%"
 SET "WEBUI_JWT_SECRET_KEY=%WEBUI_JWT_SECRET_KEY%"
 
 :: Check if WEBUI_SECRET_KEY and WEBUI_JWT_SECRET_KEY are not set
 IF "%WEBUI_SECRET_KEY%%WEBUI_JWT_SECRET_KEY%" == " " (
-    echo No WEBUI_SECRET_KEY provided
+    echo Loading WEBUI_SECRET_KEY from file, not provided as an environment variable.
 
     IF NOT EXIST "%KEY_FILE%" (
         echo Generating WEBUI_SECRET_KEY

backend/static/user-import.csv

@@ -0,0 +1 @@
+Name,Email,Password,Role

backend/utils/misc.py

@@ -38,9 +38,10 @@ def calculate_sha256_string(string):
 
 
 def validate_email_format(email: str) -> bool:
-    if not re.match(r"[^@]+@[^@]+\.[^@]+", email):
-        return False
-    return True
+    if email.endswith("@localhost"):
+        return True
+
+    return bool(re.match(r"[^@]+@[^@]+\.[^@]+", email))
 
 
 def sanitize_filename(file_name):

backend/utils/utils.py

@@ -89,6 +89,8 @@ def get_current_user(
                 status_code=status.HTTP_401_UNAUTHORIZED,
                 detail=ERROR_MESSAGES.INVALID_TOKEN,
             )
+        else:
+            Users.update_user_last_active_by_id(user.id)
         return user
     else:
         raise HTTPException(
@@ -99,11 +101,15 @@ def get_current_user(
 
 def get_current_user_by_api_key(api_key: str):
     user = Users.get_user_by_api_key(api_key)
+
     if user is None:
         raise HTTPException(
             status_code=status.HTTP_401_UNAUTHORIZED,
             detail=ERROR_MESSAGES.INVALID_TOKEN,
         )
+    else:
+        Users.update_user_last_active_by_id(user.id)
+
     return user
 
 

confirm_remove.sh

@@ -2,7 +2,12 @@
 echo "Warning: This will remove all containers and volumes, including persistent data. Do you want to continue? [Y/N]"
 read ans
 if [ "$ans" == "Y" ] || [ "$ans" == "y" ]; then
-  docker-compose down -v
+  command docker-compose 2>/dev/null
+  if [ "$?" == "0" ]; then
+    docker-compose down -v
+  else
+    docker compose down -v
+  fi
 else
   echo "Operation cancelled."
 fi

cypress.config.ts

@@ -0,0 +1,8 @@
+import { defineConfig } from 'cypress';
+
+export default defineConfig({
+	e2e: {
+		baseUrl: 'http://localhost:8080'
+	},
+	video: true
+});

cypress/e2e/chat.cy.ts

@@ -0,0 +1,78 @@
+// eslint-disable-next-line @typescript-eslint/triple-slash-reference
+/// <reference path="../support/index.d.ts" />
+
+// These tests run through the chat flow.
+describe('Settings', () => {
+	// Wait for 2 seconds after all tests to fix an issue with Cypress's video recording missing the last few frames
+	after(() => {
+		// eslint-disable-next-line cypress/no-unnecessary-waiting
+		cy.wait(2000);
+	});
+
+	beforeEach(() => {
+		// Login as the admin user
+		cy.loginAdmin();
+		// Visit the home page
+		cy.visit('/');
+	});
+
+	context('Ollama', () => {
+		it('user can select a model', () => {
+			// Click on the model selector
+			cy.get('button[aria-label="Select a model"]').click();
+			// Select the first model
+			cy.get('button[aria-label="model-item"]').first().click();
+		});
+
+		it('user can perform text chat', () => {
+			// Click on the model selector
+			cy.get('button[aria-label="Select a model"]').click();
+			// Select the first model
+			cy.get('button[aria-label="model-item"]').first().click();
+			// Type a message
+			cy.get('#chat-textarea').type('Hi, what can you do? A single sentence only please.', {
+				force: true
+			});
+			// Send the message
+			cy.get('button[type="submit"]').click();
+			// User's message should be visible
+			cy.get('.chat-user').should('exist');
+			// Wait for the response
+			cy.get('.chat-assistant', { timeout: 120_000 }) // .chat-assistant is created after the first token is received
+				.find('div[aria-label="Generation Info"]', { timeout: 120_000 }) // Generation Info is created after the stop token is received
+				.should('exist');
+		});
+
+		it('user can share chat', () => {
+			// Click on the model selector
+			cy.get('button[aria-label="Select a model"]').click();
+			// Select the first model
+			cy.get('button[aria-label="model-item"]').first().click();
+			// Type a message
+			cy.get('#chat-textarea').type('Hi, what can you do? A single sentence only please.', {
+				force: true
+			});
+			// Send the message
+			cy.get('button[type="submit"]').click();
+			// User's message should be visible
+			cy.get('.chat-user').should('exist');
+			// Wait for the response
+			cy.get('.chat-assistant', { timeout: 120_000 }) // .chat-assistant is created after the first token is received
+				.find('div[aria-label="Generation Info"]', { timeout: 120_000 }) // Generation Info is created after the stop token is received
+				.should('exist');
+			// spy on requests
+			const spy = cy.spy();
+			cy.intercept('GET', '/api/v1/chats/*', spy);
+			// Open context menu
+			cy.get('#chat-context-menu-button').click();
+			// Click share button
+			cy.get('#chat-share-button').click();
+			// Check if the share dialog is visible
+			cy.get('#copy-and-share-chat-button').should('exist');
+			cy.wrap({}, { timeout: 5000 }).should(() => {
+				// Check if the request was made twice (once for to replace chat object and once more due to change event)
+				expect(spy).to.be.callCount(2);
+			});
+		});
+	});
+});

cypress/e2e/registration.cy.ts

@@ -0,0 +1,52 @@
+// eslint-disable-next-line @typescript-eslint/triple-slash-reference
+/// <reference path="../support/index.d.ts" />
+import { adminUser } from '../support/e2e';
+
+// These tests assume the following defaults:
+// 1. No users exist in the database or that the test admin user is an admin
+// 2. Language is set to English
+// 3. The default role for new users is 'pending'
+describe('Registration and Login', () => {
+	// Wait for 2 seconds after all tests to fix an issue with Cypress's video recording missing the last few frames
+	after(() => {
+		// eslint-disable-next-line cypress/no-unnecessary-waiting
+		cy.wait(2000);
+	});
+
+	beforeEach(() => {
+		cy.visit('/');
+	});
+
+	it('should register a new user as pending', () => {
+		const userName = `Test User - ${Date.now()}`;
+		const userEmail = `cypress-${Date.now()}@example.com`;
+		// Toggle from sign in to sign up
+		cy.contains('Sign up').click();
+		// Fill out the form
+		cy.get('input[autocomplete="name"]').type(userName);
+		cy.get('input[autocomplete="email"]').type(userEmail);
+		cy.get('input[type="password"]').type('password');
+		// Submit the form
+		cy.get('button[type="submit"]').click();
+		// Wait until the user is redirected to the home page
+		cy.contains(userName);
+		// Expect the user to be pending
+		cy.contains('Check Again');
+	});
+
+	it('can login with the admin user', () => {
+		// Fill out the form
+		cy.get('input[autocomplete="email"]').type(adminUser.email);
+		cy.get('input[type="password"]').type(adminUser.password);
+		// Submit the form
+		cy.get('button[type="submit"]').click();
+		// Wait until the user is redirected to the home page
+		cy.contains(adminUser.name);
+		// Dismiss the changelog dialog if it is visible
+		cy.getAllLocalStorage().then((ls) => {
+			if (!ls['version']) {
+				cy.get('button').contains("Okay, Let's Go!").click();
+			}
+		});
+	});
+});

cypress/e2e/settings.cy.ts

@@ -0,0 +1,84 @@
+// eslint-disable-next-line @typescript-eslint/triple-slash-reference
+/// <reference path="../support/index.d.ts" />
+import { adminUser } from '../support/e2e';
+
+// These tests run through the various settings pages, ensuring that the user can interact with them as expected
+describe('Settings', () => {
+	// Wait for 2 seconds after all tests to fix an issue with Cypress's video recording missing the last few frames
+	after(() => {
+		// eslint-disable-next-line cypress/no-unnecessary-waiting
+		cy.wait(2000);
+	});
+
+	beforeEach(() => {
+		// Login as the admin user
+		cy.loginAdmin();
+		// Visit the home page
+		cy.visit('/');
+		// Click on the user menu
+		cy.get('button[aria-label="User Menu"]').click();
+		// Click on the settings link
+		cy.get('button').contains('Settings').click();
+	});
+
+	context('General', () => {
+		it('user can open the General modal and hit save', () => {
+			cy.get('button').contains('General').click();
+			cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('Connections', () => {
+		it('user can open the Connections modal and hit save', () => {
+			cy.get('button').contains('Connections').click();
+			cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('Models', () => {
+		it('user can open the Models modal', () => {
+			cy.get('button').contains('Models').click();
+		});
+	});
+
+	context('Interface', () => {
+		it('user can open the Interface modal and hit save', () => {
+			cy.get('button').contains('Interface').click();
+			cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('Audio', () => {
+		it('user can open the Audio modal and hit save', () => {
+			cy.get('button').contains('Audio').click();
+			cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('Images', () => {
+		it('user can open the Images modal and hit save', () => {
+			cy.get('button').contains('Images').click();
+			// Currently fails because the backend requires a valid URL
+			// cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('Chats', () => {
+		it('user can open the Chats modal', () => {
+			cy.get('button').contains('Chats').click();
+		});
+	});
+
+	context('Account', () => {
+		it('user can open the Account modal and hit save', () => {
+			cy.get('button').contains('Account').click();
+			cy.get('button').contains('Save').click();
+		});
+	});
+
+	context('About', () => {
+		it('user can open the About modal', () => {
+			cy.get('button').contains('About').click();
+		});
+	});
+});

cypress/support/e2e.ts

@@ -0,0 +1,73 @@
+/// <reference types="cypress" />
+
+export const adminUser = {
+	name: 'Admin User',
+	email: 'admin@example.com',
+	password: 'password'
+};
+
+const login = (email: string, password: string) => {
+	return cy.session(
+		email,
+		() => {
+			// Visit auth page
+			cy.visit('/auth');
+			// Fill out the form
+			cy.get('input[autocomplete="email"]').type(email);
+			cy.get('input[type="password"]').type(password);
+			// Submit the form
+			cy.get('button[type="submit"]').click();
+			// Wait until the user is redirected to the home page
+			cy.get('#chat-search').should('exist');
+			// Get the current version to skip the changelog dialog
+			if (localStorage.getItem('version') === null) {
+				cy.get('button').contains("Okay, Let's Go!").click();
+			}
+		},
+		{
+			validate: () => {
+				cy.request({
+					method: 'GET',
+					url: '/api/v1/auths/',
+					headers: {
+						Authorization: 'Bearer ' + localStorage.getItem('token')
+					}
+				});
+			}
+		}
+	);
+};
+
+const register = (name: string, email: string, password: string) => {
+	return cy
+		.request({
+			method: 'POST',
+			url: '/api/v1/auths/signup',
+			body: {
+				name: name,
+				email: email,
+				password: password
+			},
+			failOnStatusCode: false
+		})
+		.then((response) => {
+			expect(response.status).to.be.oneOf([200, 400]);
+		});
+};
+
+const registerAdmin = () => {
+	return register(adminUser.name, adminUser.email, adminUser.password);
+};
+
+const loginAdmin = () => {
+	return login(adminUser.email, adminUser.password);
+};
+
+Cypress.Commands.add('login', (email, password) => login(email, password));
+Cypress.Commands.add('register', (name, email, password) => register(name, email, password));
+Cypress.Commands.add('registerAdmin', () => registerAdmin());
+Cypress.Commands.add('loginAdmin', () => loginAdmin());
+
+before(() => {
+	cy.registerAdmin();
+});

cypress/support/index.d.ts

@@ -0,0 +1,11 @@
+// load the global Cypress types
+/// <reference types="cypress" />
+
+declare namespace Cypress {
+	interface Chainable {
+		login(email: string, password: string): Chainable<Element>;
+		register(name: string, email: string, password: string): Chainable<Element>;
+		registerAdmin(): Chainable<Element>;
+		loginAdmin(): Chainable<Element>;
+	}
+}

cypress/tsconfig.json

@@ -0,0 +1,7 @@
+{
+	"extends": "../tsconfig.json",
+	"compilerOptions": {
+		"inlineSourceMap": true,
+		"sourceMap": false
+	}
+}

docker-compose.api.yaml

@@ -1,5 +1,3 @@
-version: '3.8'
-
 services:
   ollama:
     # Expose Ollama API outside the container stack

docker-compose.data.yaml

@@ -1,5 +1,3 @@
-version: '3.8'
-
 services:
   ollama:
     volumes:

docker-compose.gpu.yaml

@@ -1,5 +1,3 @@
-version: '3.8'
-
 services:
   ollama:
     # GPU support

docker-compose.yaml

@@ -1,5 +1,3 @@
-version: '3.8'
-
 services:
   ollama:
     volumes:

docs/CONTRIBUTING.md

@@ -17,7 +17,7 @@ If your issue or contribution pertains directly to the core Ollama technology, p
 
 ### 🚨 Reporting Issues
 
-Noticed something off? Have an idea? Check our [Issues tab](https://github.com/open-webui/oopen-webui/issues) to see if it's already been reported or suggested. If not, feel free to open a new issue. When reporting an issue, please follow our issue templates. These templates are designed to ensure that all necessary details are provided from the start, enabling us to address your concerns more efficiently.
+Noticed something off? Have an idea? Check our [Issues tab](https://github.com/open-webui/open-webui/issues) to see if it's already been reported or suggested. If not, feel free to open a new issue. When reporting an issue, please follow our issue templates. These templates are designed to ensure that all necessary details are provided from the start, enabling us to address your concerns more efficiently.
 
 > [!IMPORTANT]
 >
@@ -54,7 +54,7 @@ Help us make Open WebUI more accessible by improving documentation, writing tuto
 
 Help us make Open WebUI available to a wider audience. In this section, we'll guide you through the process of adding new translations to the project.
 
-We use JSON files to store translations. You can find the existing translation files in the `src/lib/i18n/locales` directory. Each directory corresponds to a specific language, for example, `en-US` for English (US), `fr-FR` for French (France) and so on. You can refer to [ISO 639 Language Codes][http://www.lingoes.net/en/translator/langcode.htm] to find the appropriate code for a specific language.
+We use JSON files to store translations. You can find the existing translation files in the `src/lib/i18n/locales` directory. Each directory corresponds to a specific language, for example, `en-US` for English (US), `fr-FR` for French (France) and so on. You can refer to [ISO 639 Language Codes](http://www.lingoes.net/en/translator/langcode.htm) to find the appropriate code for a specific language.
 
 To add a new language:
 

kubernetes/helm/.helmignore

@@ -1 +0,0 @@
-values-minikube.yaml

kubernetes/helm/Chart.yaml

@@ -1,21 +0,0 @@
-apiVersion: v2
-name: open-webui
-version: 1.0.0
-appVersion: "latest"
-
-home: https://www.openwebui.com/
-icon: https://raw.githubusercontent.com/open-webui/open-webui/main/static/favicon.png
-
-description: "Open WebUI: A User-Friendly Web Interface for Chat Interactions 👋"
-keywords:
-- llm
-- chat
-- web-ui
-
-sources:
-- https://github.com/open-webui/open-webui/tree/main/kubernetes/helm
-- https://hub.docker.com/r/ollama/ollama
-- https://github.com/open-webui/open-webui/pkgs/container/open-webui
-
-annotations:
-  licenses: MIT

kubernetes/helm/README.md

@@ -0,0 +1,4 @@
+# Helm Charts
+Open WebUI Helm Charts are now hosted in a separate repo, which can be found here: https://github.com/open-webui/helm-charts 
+
+The charts are released at https://helm.openwebui.com. 

kubernetes/helm/templates/_helpers.tpl

@@ -1,51 +0,0 @@
-{{- define "open-webui.name" -}}
-{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
-{{- end -}}
-
-{{- define "ollama.name" -}}
-ollama
-{{- end -}}
-
-{{- define "ollama.url" -}}
-{{- if .Values.ollama.externalHost }}
-{{- printf .Values.ollama.externalHost }}
-{{- else }}
-{{- printf "http://%s.%s.svc.cluster.local:%d/" (include "ollama.name" .) (.Release.Namespace) (.Values.ollama.service.port | int) }}
-{{- end }}
-{{- end }}
-
-{{- define "chart.name" -}}
-{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
-{{- end }}
-
-{{- define "base.labels" -}}
-helm.sh/chart: {{ include "chart.name" . }}
-{{- if .Chart.AppVersion }}
-app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
-{{- end }}
-app.kubernetes.io/managed-by: {{ .Release.Service }}
-{{- end }}
-
-{{- define "base.selectorLabels" -}}
-app.kubernetes.io/instance: {{ .Release.Name }}
-{{- end -}}
-
-{{- define "open-webui.selectorLabels" -}}
-{{ include "base.selectorLabels" . }}
-app.kubernetes.io/component: {{ .Chart.Name }}
-{{- end }}
-
-{{- define "open-webui.labels" -}}
-{{ include "base.labels" . }}
-{{ include "open-webui.selectorLabels" . }}
-{{- end }}
-
-{{- define "ollama.selectorLabels" -}}
-{{ include "base.selectorLabels" . }}
-app.kubernetes.io/component: {{ include "ollama.name" . }}
-{{- end }}
-
-{{- define "ollama.labels" -}}
-{{ include "base.labels" . }}
-{{ include "ollama.selectorLabels" . }}
-{{- end }}

kubernetes/helm/templates/ollama-service.yaml

@@ -1,23 +0,0 @@
-{{- if not .Values.ollama.externalHost }}
-apiVersion: v1
-kind: Service
-metadata:
-  name: {{ include "ollama.name" . }}
-  labels:
-    {{- include "ollama.labels" . | nindent 4 }}
-  {{- with .Values.ollama.service.annotations }}
-  annotations:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-spec:
-  selector:
-    {{- include "ollama.selectorLabels" . | nindent 4 }}
-{{- with .Values.ollama.service }}
-  type: {{ .type }}
-  ports:
-  - protocol: TCP
-    name: http
-    port: {{ .port }}
-    targetPort: http
-{{- end }}
-{{- end }}

kubernetes/helm/templates/ollama-statefulset.yaml

@@ -1,98 +0,0 @@
-{{- if not .Values.ollama.externalHost }}
-apiVersion: apps/v1
-kind: StatefulSet
-metadata:
-  name: {{ include "ollama.name" . }}
-  labels:
-    {{- include "ollama.labels" . | nindent 4 }}
-  {{- with .Values.ollama.annotations }}
-  annotations:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-spec:
-  serviceName: {{ include "ollama.name" . }}
-  replicas: {{ .Values.ollama.replicaCount }}
-  selector:
-    matchLabels:
-      {{- include "ollama.selectorLabels" . | nindent 6 }}
-  template:
-    metadata:
-      labels:
-        {{- include "ollama.labels" . | nindent 8 }}
-      {{- with .Values.ollama.podAnnotations }}
-      annotations:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-    spec:
-      enableServiceLinks: false
-      automountServiceAccountToken: false
-      {{- with .Values.ollama.runtimeClassName }}
-      runtimeClassName: {{ . }}
-      {{- end }}
-      containers:
-      - name: {{ include "ollama.name" . }}
-        {{- with .Values.ollama.image }}
-        image: {{ .repository }}:{{ .tag }}
-        imagePullPolicy: {{ .pullPolicy }}
-        {{- end }}
-        tty: true
-        ports:
-        - name: http
-          containerPort: {{ .Values.ollama.service.containerPort }}
-        env:
-        {{- if .Values.ollama.gpu.enabled }}
-          - name: PATH
-            value: /usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
-          - name: LD_LIBRARY_PATH
-            value: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
-          - name: NVIDIA_DRIVER_CAPABILITIES
-            value: compute,utility
-        {{- end }}
-        {{- with .Values.ollama.resources }}
-        resources: {{- toYaml . | nindent 10 }}
-        {{- end }}
-        volumeMounts:
-        - name: data
-          mountPath: /root/.ollama
-      {{- with .Values.ollama.nodeSelector }}
-      nodeSelector:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-      {{- with .Values.ollama.tolerations }}
-      tolerations:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-      volumes:
-      {{- if and .Values.ollama.persistence.enabled .Values.ollama.persistence.existingClaim }}
-      - name: data
-        persistentVolumeClaim:
-          claimName: {{ .Values.ollama.persistence.existingClaim }}
-      {{- else if not .Values.ollama.persistence.enabled }}
-      - name: data
-        emptyDir: {}
-      {{- else if and .Values.ollama.persistence.enabled (not .Values.ollama.persistence.existingClaim) }}
-        []
-  volumeClaimTemplates:
-  - metadata:
-      name: data
-      labels:
-        {{- include "ollama.selectorLabels" . | nindent 8 }}
-      {{- with .Values.ollama.persistence.annotations }}
-      annotations:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-    spec:
-      accessModes:
-        {{- range .Values.ollama.persistence.accessModes }}
-        - {{ . | quote }}
-        {{- end }}
-      resources:
-        requests:
-          storage: {{ .Values.ollama.persistence.size | quote }}
-      storageClassName: {{ .Values.ollama.persistence.storageClass }}
-      {{- with .Values.ollama.persistence.selector }}
-      selector:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-      {{- end }}
-{{- end }}

kubernetes/helm/templates/webui-deployment.yaml

@@ -1,62 +0,0 @@
-apiVersion: apps/v1
-kind: Deployment
-metadata:
-  name: {{ include "open-webui.name" . }}
-  labels:
-    {{- include "open-webui.labels" . | nindent 4 }}
-  {{- with .Values.webui.annotations }}
-  annotations:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-spec:
-  replicas: {{ .Values.webui.replicaCount }}
-  selector:
-    matchLabels:
-      {{- include "open-webui.selectorLabels" . | nindent 6 }}
-  template:
-    metadata:
-      labels:
-        {{- include "open-webui.labels" . | nindent 8 }}
-      {{- with .Values.webui.podAnnotations }}
-      annotations:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-    spec:
-      enableServiceLinks: false
-      automountServiceAccountToken: false
-      containers:
-      - name: {{ .Chart.Name }}
-        {{- with .Values.webui.image }}
-        image: {{ .repository }}:{{ .tag | default $.Chart.AppVersion }}
-        imagePullPolicy: {{ .pullPolicy }}
-        {{- end }}
-        ports:
-        - name: http
-          containerPort: {{ .Values.webui.service.containerPort }}
-        {{- with .Values.webui.resources }}
-        resources: {{- toYaml . | nindent 10 }}
-        {{- end }}
-        volumeMounts:
-        - name: data
-          mountPath: /app/backend/data
-        env:
-        - name: OLLAMA_BASE_URL
-          value: {{ include "ollama.url" . | quote }}
-        tty: true
-      {{- with .Values.webui.nodeSelector }}
-      nodeSelector:
-        {{- toYaml . | nindent 8 }}
-      {{- end }}
-      volumes:
-      {{- if and .Values.webui.persistence.enabled .Values.webui.persistence.existingClaim }}
-      - name: data
-        persistentVolumeClaim:
-          claimName: {{ .Values.webui.persistence.existingClaim }}
-      {{- else if not .Values.webui.persistence.enabled }}
-      - name: data
-        emptyDir: {}
-      {{- else if and .Values.webui.persistence.enabled (not .Values.webui.persistence.existingClaim) }}
-      - name: data
-        persistentVolumeClaim:
-          claimName: {{ include "open-webui.name" . }}
-      {{- end }}

kubernetes/helm/templates/webui-ingress.yaml

@@ -1,33 +0,0 @@
-{{- if .Values.webui.ingress.enabled }}
-apiVersion: networking.k8s.io/v1
-kind: Ingress
-metadata:
-  name: {{ include "open-webui.name" . }}
-  labels:
-    {{- include "open-webui.labels" . | nindent 4 }}
-  {{- with .Values.webui.ingress.annotations }}
-  annotations:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-spec:
-  {{- with .Values.webui.ingress.class }}
-  ingressClassName: {{ . }}
-  {{- end }}
-  {{- if .Values.webui.ingress.tls }}
-  tls:
-    - hosts:
-      - {{ .Values.webui.ingress.host | quote }}
-      secretName: {{ default (printf "%s-tls" .Release.Name) .Values.webui.ingress.existingSecret }}
-  {{- end }}
-  rules:
-  - host: {{ .Values.webui.ingress.host }}
-    http:
-      paths:
-      - path: /
-        pathType: Prefix
-        backend:
-          service:
-            name: {{ include "open-webui.name" . }}
-            port:
-              name: http
-{{- end }}

kubernetes/helm/templates/webui-pvc.yaml

@@ -1,27 +0,0 @@
-{{- if and .Values.webui.persistence.enabled (not .Values.webui.persistence.existingClaim) }}
-apiVersion: v1
-kind: PersistentVolumeClaim
-metadata:
-  name: {{ include "open-webui.name" . }}
-  labels:
-    {{- include "open-webui.selectorLabels" . | nindent 4 }}
-  {{- with .Values.webui.persistence.annotations }}
-  annotations:
-    {{- toYaml . | nindent 8 }}
-  {{- end }}
-spec:
-  accessModes:
-    {{- range .Values.webui.persistence.accessModes }}
-    - {{ . | quote }}
-    {{- end }}
-  resources:
-    requests:
-      storage: {{ .Values.webui.persistence.size }}
-  {{- if .Values.webui.persistence.storageClass }}
-  storageClassName: {{ .Values.webui.persistence.storageClass }}
-  {{- end }}
-  {{- with .Values.webui.persistence.selector }}
-  selector:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-{{- end }}

kubernetes/helm/templates/webui-service.yaml

@@ -1,29 +0,0 @@
-apiVersion: v1
-kind: Service
-metadata:
-  name: {{ include "open-webui.name" . }}
-  labels:
-    {{- include "open-webui.labels" . | nindent 4 }}
-    {{- with .Values.webui.service.labels }}
-    {{- toYaml . | nindent 4 }}
-    {{- end }}
-  {{- with .Values.webui.service.annotations }}
-  annotations:
-    {{- toYaml . | nindent 4 }}
-  {{- end }}
-spec:
-  selector:
-    {{- include "open-webui.selectorLabels" . | nindent 4 }}
-  type: {{ .Values.webui.service.type | default "ClusterIP" }}
-  ports:
-  - protocol: TCP
-    name: http
-    port: {{ .Values.webui.service.port }}
-    targetPort: http
-    {{- if .Values.webui.service.nodePort }}
-    nodePort: {{ .Values.webui.service.nodePort | int }}
-    {{- end }}
-  {{- if .Values.webui.service.loadBalancerClass }}
-  loadBalancerClass: {{ .Values.webui.service.loadBalancerClass | quote }}
-  {{- end }}
-

kubernetes/helm/values-minikube.yaml

@@ -1,27 +0,0 @@
-ollama:
-  resources:
-    requests:
-      cpu: "2000m"
-      memory: "2Gi"
-    limits:
-      cpu: "4000m"
-      memory: "4Gi"
-      nvidia.com/gpu: "0"
-  service:
-    type: ClusterIP
-  gpu:
-    enabled: false
-
-webui:
-  resources:
-    requests:
-      cpu: "500m"
-      memory: "500Mi"
-    limits:
-      cpu: "1000m"
-      memory: "1Gi"
-  ingress:
-    enabled: true
-    host: open-webui.minikube.local
-  service:
-    type: NodePort

kubernetes/helm/values.yaml

@@ -1,75 +0,0 @@
-nameOverride: ""
-
-ollama:
-  externalHost: ""
-  annotations: {}
-  podAnnotations: {}
-  replicaCount: 1
-  image:
-    repository: ollama/ollama
-    tag: latest
-    pullPolicy: Always
-  resources: {}
-  persistence:
-    enabled: true
-    size: 30Gi
-    existingClaim: ""
-    accessModes:
-    - ReadWriteOnce
-    storageClass: ""
-    selector: {}
-    annotations: {}
-  nodeSelector: {}
-  # -- If using a special runtime container such as nvidia, set it here.
-  runtimeClassName: ""
-  tolerations:
-  - key: nvidia.com/gpu
-    operator: Exists
-    effect: NoSchedule
-  service:
-    type: ClusterIP
-    annotations: {}
-    port: 80
-    containerPort: 11434
-  gpu:
-    # -- Enable additional ENV values to help Ollama discover GPU usage
-    enabled: false
-
-webui:
-  annotations: {}
-  podAnnotations: {}
-  replicaCount: 1
-  image:
-    repository: ghcr.io/open-webui/open-webui
-    tag: ""
-    pullPolicy: Always
-  resources: {}
-  ingress:
-    enabled: false
-    class: ""
-    # -- Use appropriate annotations for your Ingress controller, e.g., for NGINX:
-    # nginx.ingress.kubernetes.io/rewrite-target: /
-    annotations: {}
-    host: ""
-    tls: false
-    existingSecret: ""
-  persistence:
-    enabled: true
-    size: 2Gi
-    existingClaim: ""
-    # -- If using multiple replicas, you must update accessModes to ReadWriteMany
-    accessModes:
-    - ReadWriteOnce
-    storageClass: ""
-    selector: {}
-    annotations: {}
-  nodeSelector: {}
-  tolerations: []
-  service:
-    type: ClusterIP
-    annotations: {}
-    port: 80
-    containerPort: 8080
-    nodePort: ""
-    labels: {}
-    loadBalancerClass: "" 

package.json

@@ -1,10 +1,10 @@
 {
 	"name": "open-webui",
-	"version": "0.1.118",
+	"version": "0.1.125",
 	"private": true,
 	"scripts": {
-		"dev": "vite dev --host",
-		"build": "vite build",
+		"dev": "npm run pyodide:fetch && vite dev --host",
+		"build": "npm run pyodide:fetch && vite build",
 		"preview": "vite preview",
 		"check": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json",
 		"check:watch": "svelte-kit sync && svelte-check --tsconfig ./tsconfig.json --watch",
@@ -12,9 +12,12 @@
 		"lint:frontend": "eslint . --fix",
 		"lint:types": "npm run check",
 		"lint:backend": "pylint backend/",
-		"format": "prettier --plugin-search-dir --write '**/*.{js,ts,svelte,css,md,html,json}'",
+		"format": "prettier --plugin-search-dir --write \"**/*.{js,ts,svelte,css,md,html,json}\"",
 		"format:backend": "black . --exclude \"/venv/\"",
-		"i18n:parse": "i18next --config i18next-parser.config.ts && prettier --write 'src/lib/i18n/**/*.{js,json}'"
+		"i18n:parse": "i18next --config i18next-parser.config.ts && prettier --write \"src/lib/i18n/**/*.{js,json}\"",
+		"cy:open": "cypress open",
+		"test:frontend": "vitest",
+		"pyodide:fetch": "node scripts/prepare-pyodide.js"
 	},
 	"devDependencies": {
 		"@sveltejs/adapter-auto": "^2.0.0",
@@ -25,8 +28,10 @@
 		"@typescript-eslint/eslint-plugin": "^6.17.0",
 		"@typescript-eslint/parser": "^6.17.0",
 		"autoprefixer": "^10.4.16",
+		"cypress": "^13.8.1",
 		"eslint": "^8.56.0",
 		"eslint-config-prettier": "^8.5.0",
+		"eslint-plugin-cypress": "^3.0.2",
 		"eslint-plugin-svelte": "^2.30.0",
 		"i18next-parser": "^8.13.0",
 		"postcss": "^8.4.31",
@@ -38,14 +43,17 @@
 		"tailwindcss": "^3.3.3",
 		"tslib": "^2.4.1",
 		"typescript": "^5.0.0",
-		"vite": "^4.4.2"
+		"vite": "^4.4.2",
+		"vitest": "^1.6.0"
 	},
 	"type": "module",
 	"dependencies": {
+		"@pyscript/core": "^0.4.32",
 		"@sveltejs/adapter-node": "^1.3.1",
 		"async": "^3.2.5",
 		"bits-ui": "^0.19.7",
 		"dayjs": "^1.11.10",
+		"eventsource-parser": "^1.1.2",
 		"file-saver": "^2.0.5",
 		"highlight.js": "^11.9.0",
 		"i18next": "^23.10.0",
@@ -53,9 +61,9 @@
 		"i18next-resources-to-backend": "^1.2.0",
 		"idb": "^7.1.1",
 		"js-sha256": "^0.10.1",
-		"jspdf": "^2.5.1",
 		"katex": "^0.16.9",
 		"marked": "^9.1.0",
+		"pyodide": "^0.26.0-alpha.4",
 		"svelte-sonner": "^0.3.19",
 		"tippy.js": "^6.3.7",
 		"uuid": "^9.0.1"

run-compose.sh

@@ -82,6 +82,7 @@ usage() {
     echo "Examples:"
     echo "  $0 --drop"
     echo "  $0 --enable-gpu[count=1]"
+    echo "  $0 --enable-gpu[count=all]"
     echo "  $0 --enable-api[port=11435]"
     echo "  $0 --enable-gpu[count=1] --enable-api[port=12345] --webui[port=3000]"
     echo "  $0 --enable-gpu[count=1] --enable-api[port=12345] --webui[port=3000] --data[folder=./ollama-data]"
@@ -160,7 +161,7 @@ else
     if [[ $enable_gpu == true ]]; then
         # Validate and process command-line arguments
         if [[ -n $gpu_count ]]; then
-            if ! [[ $gpu_count =~ ^[0-9]+$ ]]; then
+            if ! [[ $gpu_count =~ ^([0-9]+|all)$ ]]; then
                 echo "Invalid GPU count: $gpu_count"
                 exit 1
             fi

scripts/prepare-pyodide.js

@@ -0,0 +1,39 @@
+const packages = [
+	'requests',
+	'beautifulsoup4',
+	'numpy',
+	'pandas',
+	'matplotlib',
+	'scikit-learn',
+	'scipy',
+	'regex',
+	'seaborn'
+];
+
+import { loadPyodide } from 'pyodide';
+import { writeFile, copyFile, readdir } from 'fs/promises';
+
+async function downloadPackages() {
+	console.log('Setting up pyodide + micropip');
+	const pyodide = await loadPyodide({
+		packageCacheDir: 'static/pyodide'
+	});
+	await pyodide.loadPackage('micropip');
+	const micropip = pyodide.pyimport('micropip');
+	console.log('Downloading Pyodide packages:', packages);
+	await micropip.install(packages);
+	console.log('Pyodide packages downloaded, freezing into lock file');
+	const lockFile = await micropip.freeze();
+	await writeFile('static/pyodide/pyodide-lock.json', lockFile);
+}
+
+async function copyPyodide() {
+	console.log('Copying Pyodide files into static directory');
+	// Copy all files from node_modules/pyodide to static/pyodide
+	for await (const entry of await readdir('node_modules/pyodide')) {
+		await copyFile(`node_modules/pyodide/${entry}`, `static/pyodide/${entry}`);
+	}
+}
+
+await downloadPackages();
+await copyPyodide();

src/app.css

@@ -82,3 +82,32 @@ select {
 .katex-mathml {
 	display: none;
 }
+
+.scrollbar-hidden:active::-webkit-scrollbar-thumb,
+.scrollbar-hidden:focus::-webkit-scrollbar-thumb,
+.scrollbar-hidden:hover::-webkit-scrollbar-thumb {
+	visibility: visible;
+}
+.scrollbar-hidden::-webkit-scrollbar-thumb {
+	visibility: hidden;
+}
+
+.scrollbar-none::-webkit-scrollbar {
+	display: none; /* for Chrome, Safari and Opera */
+}
+
+.scrollbar-none {
+	-ms-overflow-style: none; /* IE and Edge */
+	scrollbar-width: none; /* Firefox */
+}
+
+input::-webkit-outer-spin-button,
+input::-webkit-inner-spin-button {
+	/* display: none; <- Crashes Chrome on hover */
+	-webkit-appearance: none;
+	margin: 0; /* <-- Apparently some margin are still there even though it's hidden */
+}
+
+input[type='number'] {
+	-moz-appearance: textfield; /* Firefox */
+}

src/app.html

@@ -3,9 +3,16 @@
 	<head>
 		<meta charset="utf-8" />
 		<link rel="icon" href="%sveltekit.assets%/favicon.png" />
-		<link rel="manifest" href="%sveltekit.assets%/manifest.json" />
+		<link rel="manifest" href="%sveltekit.assets%/manifest.json" crossorigin="use-credentials" />
 		<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1" />
 		<meta name="robots" content="noindex,nofollow" />
+		<link
+			rel="search"
+			type="application/opensearchdescription+xml"
+			title="Open WebUI"
+			href="/opensearch.xml"
+		/>
+
 		<script>
 			// On page load or when changing themes, best to add inline in `head` to avoid FOUC
 			(() => {
@@ -43,9 +50,46 @@
 			})();
 		</script>
 
+		<title>Open WebUI</title>
+
 		%sveltekit.head%
 	</head>
 	<body data-sveltekit-preload-data="hover">
 		<div style="display: contents">%sveltekit.body%</div>
+
+		<div
+			id="splash-screen"
+			style="
+				position: fixed;
+				z-index: 100;
+				background: #fff;
+				top: 0;
+				left: 0;
+				width: 100%;
+				height: 100%;
+			"
+		>
+			<style type="text/css" nonce="">
+				html {
+					overflow-y: scroll !important;
+				}
+			</style>
+
+			<img
+				style="
+					position: absolute;
+					width: 6rem;
+					height: 6rem;
+					top: 46%;
+					left: 50%;
+					margin: -40px 0 0 -40px;
+				"
+				src="/logo.svg"
+			/>
+
+			<!-- <span style="position: absolute; bottom: 32px; left: 50%; margin: -36px 0 0 -36px">
+				Footer content
+			</span> -->
+		</div>
 	</body>
 </html>

src/lib/apis/audio/index.ts

@@ -1,11 +1,75 @@
 import { AUDIO_API_BASE_URL } from '$lib/constants';
 
+export const getAudioConfig = async (token: string) => {
+	let error = null;
+
+	const res = await fetch(`${AUDIO_API_BASE_URL}/config`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+type OpenAIConfigForm = {
+	url: string;
+	key: string;
+	model: string;
+	speaker: string;
+};
+
+export const updateAudioConfig = async (token: string, payload: OpenAIConfigForm) => {
+	let error = null;
+
+	const res = await fetch(`${AUDIO_API_BASE_URL}/config/update`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			...payload
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const transcribeAudio = async (token: string, file: File) => {
 	const data = new FormData();
 	data.append('file', file);
 
 	let error = null;
-	const res = await fetch(`${AUDIO_API_BASE_URL}/transcribe`, {
+	const res = await fetch(`${AUDIO_API_BASE_URL}/transcriptions`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -29,3 +93,41 @@ export const transcribeAudio = async (token: string, file: File) => {
 
 	return res;
 };
+
+export const synthesizeOpenAISpeech = async (
+	token: string = '',
+	speaker: string = 'alloy',
+	text: string = '',
+	model: string = 'tts-1'
+) => {
+	let error = null;
+
+	const res = await fetch(`${AUDIO_API_BASE_URL}/speech`, {
+		method: 'POST',
+		headers: {
+			Authorization: `Bearer ${token}`,
+			'Content-Type': 'application/json'
+		},
+		body: JSON.stringify({
+			model: model,
+			input: text,
+			voice: speaker
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res;
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};

src/lib/apis/auths/index.ts

@@ -95,6 +95,45 @@ export const userSignUp = async (
 	return res;
 };
 
+export const addUser = async (
+	token: string,
+	name: string,
+	email: string,
+	password: string,
+	role: string = 'pending'
+) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/auths/add`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		},
+		body: JSON.stringify({
+			name: name,
+			email: email,
+			password: password,
+			role: role
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const updateUserProfile = async (token: string, name: string, profileImageUrl: string) => {
 	let error = null;
 

src/lib/apis/chats/index.ts

@@ -1,4 +1,5 @@
 import { WEBUI_API_BASE_URL } from '$lib/constants';
+import { getTimeRange } from '$lib/utils';
 
 export const createNewChat = async (token: string, chat: object) => {
 	let error = null;
@@ -59,6 +60,74 @@ export const getChatList = async (token: string = '') => {
 		throw error;
 	}
 
+	return res.map((chat) => ({
+		...chat,
+		time_range: getTimeRange(chat.updated_at)
+	}));
+};
+
+export const getChatListByUserId = async (token: string = '', userId: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/list/user/${userId}`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res.map((chat) => ({
+		...chat,
+		time_range: getTimeRange(chat.updated_at)
+	}));
+};
+
+export const getArchivedChatList = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/archived`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
 	return res;
 };
 
@@ -158,13 +227,16 @@ export const getAllChatTags = async (token: string) => {
 export const getChatListByTagName = async (token: string = '', tagName: string) => {
 	let error = null;
 
-	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/tags/tag/${tagName}`, {
-		method: 'GET',
+	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/tags`, {
+		method: 'POST',
 		headers: {
 			Accept: 'application/json',
 			'Content-Type': 'application/json',
 			...(token && { authorization: `Bearer ${token}` })
-		}
+		},
+		body: JSON.stringify({
+			name: tagName
+		})
 	})
 		.then(async (res) => {
 			if (!res.ok) throw await res.json();
@@ -183,7 +255,10 @@ export const getChatListByTagName = async (token: string = '', tagName: string)
 		throw error;
 	}
 
-	return res;
+	return res.map((chat) => ({
+		...chat,
+		time_range: getTimeRange(chat.updated_at)
+	}));
 };
 
 export const getChatById = async (token: string, id: string) => {
@@ -282,6 +357,38 @@ export const shareChatById = async (token: string, id: string) => {
 	return res;
 };
 
+export const archiveChatById = async (token: string, id: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/chats/${id}/archive`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const deleteSharedChatById = async (token: string, id: string) => {
 	let error = null;
 

src/lib/apis/images/index.ts

@@ -72,10 +72,10 @@ export const updateImageGenerationConfig = async (
 	return res;
 };
 
-export const getOpenAIKey = async (token: string = '') => {
+export const getOpenAIConfig = async (token: string = '') => {
 	let error = null;
 
-	const res = await fetch(`${IMAGES_API_BASE_URL}/key`, {
+	const res = await fetch(`${IMAGES_API_BASE_URL}/openai/config`, {
 		method: 'GET',
 		headers: {
 			Accept: 'application/json',
@@ -101,13 +101,13 @@ export const getOpenAIKey = async (token: string = '') => {
 		throw error;
 	}
 
-	return res.OPENAI_API_KEY;
+	return res;
 };
 
-export const updateOpenAIKey = async (token: string = '', key: string) => {
+export const updateOpenAIConfig = async (token: string = '', url: string, key: string) => {
 	let error = null;
 
-	const res = await fetch(`${IMAGES_API_BASE_URL}/key/update`, {
+	const res = await fetch(`${IMAGES_API_BASE_URL}/openai/config/update`, {
 		method: 'POST',
 		headers: {
 			Accept: 'application/json',
@@ -115,6 +115,7 @@ export const updateOpenAIKey = async (token: string = '', key: string) => {
 			...(token && { authorization: `Bearer ${token}` })
 		},
 		body: JSON.stringify({
+			url: url,
 			key: key
 		})
 	})
@@ -136,7 +137,7 @@ export const updateOpenAIKey = async (token: string = '', key: string) => {
 		throw error;
 	}
 
-	return res.OPENAI_API_KEY;
+	return res;
 };
 
 export const getImageGenerationEngineUrls = async (token: string = '') => {

src/lib/apis/memories/index.ts

@@ -0,0 +1,155 @@
+import { WEBUI_API_BASE_URL } from '$lib/constants';
+
+export const getMemories = async (token: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/memories`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const addNewMemory = async (token: string, content: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/memories/add`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			content: content
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const queryMemory = async (token: string, content: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/memories/query`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			content: content
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const deleteMemoryById = async (token: string, id: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/memories/${id}`, {
+		method: 'DELETE',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const deleteMemoriesByUserId = async (token: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/memories/user`, {
+		method: 'DELETE',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.then((json) => {
+			return json;
+		})
+		.catch((err) => {
+			error = err.detail;
+
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};

src/lib/apis/ollama/index.ts

@@ -1,4 +1,5 @@
 import { OLLAMA_API_BASE_URL } from '$lib/constants';
+import { promptTemplate } from '$lib/utils';
 
 export const getOllamaUrls = async (token: string = '') => {
 	let error = null;
@@ -144,7 +145,7 @@ export const generateTitle = async (
 ) => {
 	let error = null;
 
-	template = template.replace(/{{prompt}}/g, prompt);
+	template = promptTemplate(template, prompt);
 
 	console.log(template);
 
@@ -158,7 +159,11 @@ export const generateTitle = async (
 		body: JSON.stringify({
 			model: model,
 			prompt: template,
-			stream: false
+			stream: false,
+			options: {
+				// Restrict the number of tokens generated to 50
+				num_predict: 50
+			}
 		})
 	})
 		.then(async (res) => {
@@ -177,7 +182,7 @@ export const generateTitle = async (
 		throw error;
 	}
 
-	return res?.response ?? 'New Chat';
+	return res?.response.replace(/["']/g, '') ?? 'New Chat';
 };
 
 export const generatePrompt = async (token: string = '', model: string, conversation: string) => {
@@ -219,6 +224,32 @@ export const generatePrompt = async (token: string = '', model: string, conversa
 	return res;
 };
 
+export const generateEmbeddings = async (token: string = '', model: string, text: string) => {
+	let error = null;
+
+	const res = await fetch(`${OLLAMA_API_BASE_URL}/api/embeddings`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			model: model,
+			prompt: text
+		})
+	}).catch((err) => {
+		error = err;
+		return null;
+	});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const generateTextCompletion = async (token: string = '', model: string, text: string) => {
 	let error = null;
 

src/lib/apis/openai/index.ts

@@ -1,4 +1,72 @@
 import { OPENAI_API_BASE_URL } from '$lib/constants';
+import { promptTemplate } from '$lib/utils';
+
+export const getOpenAIConfig = async (token: string = '') => {
+	let error = null;
+
+	const res = await fetch(`${OPENAI_API_BASE_URL}/config`, {
+		method: 'GET',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			if ('detail' in err) {
+				error = err.detail;
+			} else {
+				error = 'Server connection failed';
+			}
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const updateOpenAIConfig = async (token: string = '', enable_openai_api: boolean) => {
+	let error = null;
+
+	const res = await fetch(`${OPENAI_API_BASE_URL}/config/update`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			...(token && { authorization: `Bearer ${token}` })
+		},
+		body: JSON.stringify({
+			enable_openai_api: enable_openai_api
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			if ('detail' in err) {
+				error = err.detail;
+			} else {
+				error = 'Server connection failed';
+			}
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
 
 export const getOpenAIUrls = async (token: string = '') => {
 	let error = null;
@@ -210,10 +278,12 @@ export const generateOpenAIChatCompletion = async (
 	token: string = '',
 	body: object,
 	url: string = OPENAI_API_BASE_URL
-) => {
+): Promise<[Response | null, AbortController]> => {
+	const controller = new AbortController();
 	let error = null;
 
 	const res = await fetch(`${url}/chat/completions`, {
+		signal: controller.signal,
 		method: 'POST',
 		headers: {
 			Authorization: `Bearer ${token}`,
@@ -230,13 +300,14 @@ export const generateOpenAIChatCompletion = async (
 		throw error;
 	}
 
-	return res;
+	return [res, controller];
 };
 
 export const synthesizeOpenAISpeech = async (
 	token: string = '',
 	speaker: string = 'alloy',
-	text: string = ''
+	text: string = '',
+	model: string = 'tts-1'
 ) => {
 	let error = null;
 
@@ -247,7 +318,7 @@ export const synthesizeOpenAISpeech = async (
 			'Content-Type': 'application/json'
 		},
 		body: JSON.stringify({
-			model: 'tts-1',
+			model: model,
 			input: text,
 			voice: speaker
 		})
@@ -273,7 +344,7 @@ export const generateTitle = async (
 ) => {
 	let error = null;
 
-	template = template.replace(/{{prompt}}/g, prompt);
+	template = promptTemplate(template, prompt);
 
 	console.log(template);
 
@@ -292,7 +363,9 @@ export const generateTitle = async (
 					content: template
 				}
 			],
-			stream: false
+			stream: false,
+			// Restricting the max tokens to 50 to avoid long titles
+			max_tokens: 50
 		})
 	})
 		.then(async (res) => {
@@ -311,5 +384,5 @@ export const generateTitle = async (
 		throw error;
 	}
 
-	return res?.choices[0]?.message?.content ?? 'New Chat';
+	return res?.choices[0]?.message?.content.replace(/["']/g, '') ?? 'New Chat';
 };

src/lib/apis/rag/index.ts

@@ -32,9 +32,16 @@ type ChunkConfigForm = {
 	chunk_overlap: number;
 };
 
+type YoutubeConfigForm = {
+	language: string[];
+	translation?: string | null;
+};
+
 type RAGConfigForm = {
-	pdf_extract_images: boolean;
-	chunk: ChunkConfigForm;
+	pdf_extract_images?: boolean;
+	chunk?: ChunkConfigForm;
+	web_loader_ssl_verification?: boolean;
+	youtube?: YoutubeConfigForm;
 };
 
 export const updateRAGConfig = async (token: string, payload: RAGConfigForm) => {
@@ -123,6 +130,7 @@ export const getQuerySettings = async (token: string) => {
 
 type QuerySettings = {
 	k: number | null;
+	r: number | null;
 	template: string | null;
 };
 
@@ -220,6 +228,37 @@ export const uploadWebToVectorDB = async (token: string, collection_name: string
 	return res;
 };
 
+export const uploadYoutubeTranscriptionToVectorDB = async (token: string, url: string) => {
+	let error = null;
+
+	const res = await fetch(`${RAG_API_BASE_URL}/youtube`, {
+		method: 'POST',
+		headers: {
+			Accept: 'application/json',
+			'Content-Type': 'application/json',
+			authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			url: url
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			error = err.detail;
+			console.log(err);
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const queryDoc = async (
 	token: string,
 	collection_name: string,
@@ -346,10 +385,10 @@ export const resetVectorDB = async (token: string) => {
 	return res;
 };
 
-export const getEmbeddingModel = async (token: string) => {
+export const getEmbeddingConfig = async (token: string) => {
 	let error = null;
 
-	const res = await fetch(`${RAG_API_BASE_URL}/embedding/model`, {
+	const res = await fetch(`${RAG_API_BASE_URL}/embedding`, {
 		method: 'GET',
 		headers: {
 			'Content-Type': 'application/json',
@@ -373,14 +412,82 @@ export const getEmbeddingModel = async (token: string) => {
 	return res;
 };
 
+type OpenAIConfigForm = {
+	key: string;
+	url: string;
+};
+
 type EmbeddingModelUpdateForm = {
+	openai_config?: OpenAIConfigForm;
+	embedding_engine: string;
 	embedding_model: string;
 };
 
-export const updateEmbeddingModel = async (token: string, payload: EmbeddingModelUpdateForm) => {
+export const updateEmbeddingConfig = async (token: string, payload: EmbeddingModelUpdateForm) => {
 	let error = null;
 
-	const res = await fetch(`${RAG_API_BASE_URL}/embedding/model/update`, {
+	const res = await fetch(`${RAG_API_BASE_URL}/embedding/update`, {
+		method: 'POST',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		},
+		body: JSON.stringify({
+			...payload
+		})
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+export const getRerankingConfig = async (token: string) => {
+	let error = null;
+
+	const res = await fetch(`${RAG_API_BASE_URL}/reranking`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
+type RerankingModelUpdateForm = {
+	reranking_model: string;
+};
+
+export const updateRerankingConfig = async (token: string, payload: RerankingModelUpdateForm) => {
+	let error = null;
+
+	const res = await fetch(`${RAG_API_BASE_URL}/reranking/update`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'application/json',

src/lib/apis/streaming/index.ts

@@ -0,0 +1,102 @@
+import { EventSourceParserStream } from 'eventsource-parser/stream';
+import type { ParsedEvent } from 'eventsource-parser';
+
+type TextStreamUpdate = {
+	done: boolean;
+	value: string;
+	// eslint-disable-next-line @typescript-eslint/no-explicit-any
+	citations?: any;
+	// eslint-disable-next-line @typescript-eslint/no-explicit-any
+	error?: any;
+};
+
+// createOpenAITextStream takes a responseBody with a SSE response,
+// and returns an async generator that emits delta updates with large deltas chunked into random sized chunks
+export async function createOpenAITextStream(
+	responseBody: ReadableStream<Uint8Array>,
+	splitLargeDeltas: boolean
+): Promise<AsyncGenerator<TextStreamUpdate>> {
+	const eventStream = responseBody
+		.pipeThrough(new TextDecoderStream())
+		.pipeThrough(new EventSourceParserStream())
+		.getReader();
+	let iterator = openAIStreamToIterator(eventStream);
+	if (splitLargeDeltas) {
+		iterator = streamLargeDeltasAsRandomChunks(iterator);
+	}
+	return iterator;
+}
+
+async function* openAIStreamToIterator(
+	reader: ReadableStreamDefaultReader<ParsedEvent>
+): AsyncGenerator<TextStreamUpdate> {
+	while (true) {
+		const { value, done } = await reader.read();
+		if (done) {
+			yield { done: true, value: '' };
+			break;
+		}
+		if (!value) {
+			continue;
+		}
+		const data = value.data;
+		if (data.startsWith('[DONE]')) {
+			yield { done: true, value: '' };
+			break;
+		}
+
+		try {
+			const parsedData = JSON.parse(data);
+			console.log(parsedData);
+
+			if (parsedData.error) {
+				yield { done: true, value: '', error: parsedData.error };
+				break;
+			}
+
+			if (parsedData.citations) {
+				yield { done: false, value: '', citations: parsedData.citations };
+				continue;
+			}
+
+			yield { done: false, value: parsedData.choices?.[0]?.delta?.content ?? '' };
+		} catch (e) {
+			console.error('Error extracting delta from SSE event:', e);
+		}
+	}
+}
+
+// streamLargeDeltasAsRandomChunks will chunk large deltas (length > 5) into random sized chunks between 1-3 characters
+// This is to simulate a more fluid streaming, even though some providers may send large chunks of text at once
+async function* streamLargeDeltasAsRandomChunks(
+	iterator: AsyncGenerator<TextStreamUpdate>
+): AsyncGenerator<TextStreamUpdate> {
+	for await (const textStreamUpdate of iterator) {
+		if (textStreamUpdate.done) {
+			yield textStreamUpdate;
+			return;
+		}
+		if (textStreamUpdate.citations) {
+			yield textStreamUpdate;
+			continue;
+		}
+		let content = textStreamUpdate.value;
+		if (content.length < 5) {
+			yield { done: false, value: content };
+			continue;
+		}
+		while (content != '') {
+			const chunkSize = Math.min(Math.floor(Math.random() * 3) + 1, content.length);
+			const chunk = content.slice(0, chunkSize);
+			yield { done: false, value: chunk };
+			// Do not sleep if the tab is hidden
+			// Timers are throttled to 1s in hidden tabs
+			if (document?.visibilityState !== 'hidden') {
+				await sleep(5);
+			}
+			content = content.slice(chunkSize);
+		}
+	}
+}
+
+const sleep = (ms: number) => new Promise((resolve) => setTimeout(resolve, ms));

src/lib/apis/users/index.ts

@@ -115,6 +115,33 @@ export const getUsers = async (token: string) => {
 	return res ? res : [];
 };
 
+export const getUserById = async (token: string, userId: string) => {
+	let error = null;
+
+	const res = await fetch(`${WEBUI_API_BASE_URL}/users/${userId}`, {
+		method: 'GET',
+		headers: {
+			'Content-Type': 'application/json',
+			Authorization: `Bearer ${token}`
+		}
+	})
+		.then(async (res) => {
+			if (!res.ok) throw await res.json();
+			return res.json();
+		})
+		.catch((err) => {
+			console.log(err);
+			error = err.detail;
+			return null;
+		});
+
+	if (error) {
+		throw error;
+	}
+
+	return res;
+};
+
 export const deleteUserById = async (token: string, userId: string) => {
 	let error = null;
 

src/lib/apis/utils/index.ts

@@ -83,9 +83,9 @@ export const downloadDatabase = async (token: string) => {
 			Authorization: `Bearer ${token}`
 		}
 	})
-		.then((response) => {
+		.then(async (response) => {
 			if (!response.ok) {
-				throw new Error('Network response was not ok');
+				throw await response.json();
 			}
 			return response.blob();
 		})
@@ -100,7 +100,11 @@ export const downloadDatabase = async (token: string) => {
 		})
 		.catch((err) => {
 			console.log(err);
-			error = err;
+			error = err.detail;
 			return null;
 		});
+
+	if (error) {
+		throw error;
+	}
 };

src/lib/components/ChangelogModal.svelte

@@ -22,7 +22,7 @@
 </script>
 
 <Modal bind:show>
-	<div class="px-5 py-4 dark:text-gray-300">
+	<div class="px-5 pt-4 dark:text-gray-300 text-gray-700">
 		<div class="flex justify-between items-start">
 			<div class="text-xl font-bold">
 				{$i18n.t('What’s New in')}
@@ -32,6 +32,7 @@
 			<button
 				class="self-center"
 				on:click={() => {
+					localStorage.version = $config.version;
 					show = false;
 				}}
 			>
@@ -56,10 +57,8 @@
 		</div>
 	</div>
 
-	<hr class=" dark:border-gray-800" />
-
-	<div class=" w-full p-4 px-5">
-		<div class=" overflow-y-scroll max-h-80">
+	<div class=" w-full p-4 px-5 text-gray-700 dark:text-gray-100">
+		<div class=" overflow-y-scroll max-h-80 scrollbar-hidden">
 			<div class="mb-3">
 				{#if changelog}
 					{#each Object.keys(changelog) as version}
@@ -109,7 +108,7 @@
 					localStorage.version = $config.version;
 					show = false;
 				}}
-				class=" px-4 py-2 bg-emerald-600 hover:bg-emerald-700 text-gray-100 transition rounded"
+				class=" px-4 py-2 bg-emerald-700 hover:bg-emerald-800 text-gray-100 transition rounded-lg"
 			>
 				<span class="relative">{$i18n.t("Okay, Let's Go!")}</span>
 			</button>