chore(main): release 0.6.1 (#2041)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

.docker/router.yml

@@ -0,0 +1,16 @@
+http:
+  services:
+    ollama:
+      loadBalancer:
+        healthCheck:
+          interval: 5s
+          path: /
+        servers:
+          - url: http://ollama-cpu:11434
+          - url: http://ollama-cuda:11434
+          - url: http://host.docker.internal:11434
+
+  routers:
+    ollama-router:
+      rule: "PathPrefix(`/`)"
+      service: ollama

.dockerignore

@@ -0,0 +1,12 @@
+.venv
+models
+.github
+.vscode
+.DS_Store
+.mypy_cache
+.ruff_cache
+local_data
+terraform
+tests
+Dockerfile
+Dockerfile.*

.github/ISSUE_TEMPLATE/bug.yml

@@ -0,0 +1,105 @@
+name: Bug Report
+description: Report a bug or issue with the project.
+title: "[BUG] "
+labels: ["bug"]
+
+body:
+  - type: markdown
+    attributes:
+      value: |
+        **Please describe the bug you encountered.**
+
+  - type: checkboxes
+    id: pre-check
+    attributes:
+      label: Pre-check
+      description: Please confirm that you have searched for duplicate issues before creating this one.
+      options:
+        - label: I have searched the existing issues and none cover this bug.
+          required: true
+
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      description: Provide a detailed description of the bug.
+      placeholder: "Detailed description of the bug"
+    validations:
+      required: true
+
+  - type: textarea
+    id: steps
+    attributes:
+      label: Steps to Reproduce
+      description: Provide the steps to reproduce the bug.
+      placeholder: "1. Step one\n2. Step two\n3. Step three"
+    validations:
+      required: true
+
+  - type: input
+    id: expected
+    attributes:
+      label: Expected Behavior
+      description: Describe what you expected to happen.
+      placeholder: "Expected behavior"
+    validations:
+      required: true
+
+  - type: input
+    id: actual
+    attributes:
+      label: Actual Behavior
+      description: Describe what actually happened.
+      placeholder: "Actual behavior"
+    validations:
+      required: true
+
+  - type: input
+    id: environment
+    attributes:
+      label: Environment
+      description: Provide details about your environment (e.g., OS, GPU, profile, etc.).
+      placeholder: "Environment details"
+    validations:
+      required: true
+
+  - type: input
+    id: additional
+    attributes:
+      label: Additional Information
+      description: Provide any additional information that may be relevant (e.g., logs, screenshots).
+      placeholder: "Any additional information that may be relevant"
+
+  - type: input
+    id: version
+    attributes:
+      label: Version
+      description: Provide the version of the project where you encountered the bug.
+      placeholder: "Version number"
+
+  - type: markdown
+    attributes:
+      value: |
+        **Please ensure the following setup checklist has been reviewed before submitting the bug report.**
+
+  - type: checkboxes
+    id: general-setup-checklist
+    attributes:
+      label: Setup Checklist
+      description: Verify the following general aspects of your setup.
+      options:
+        - label: Confirm that you have followed the installation instructions in the project’s documentation.
+        - label: Check that you are using the latest version of the project.
+        - label: Verify disk space availability for model storage and data processing.
+        - label: Ensure that you have the necessary permissions to run the project.
+
+  - type: checkboxes
+    id: nvidia-setup-checklist
+    attributes:
+      label: NVIDIA GPU Setup Checklist
+      description: Verify the following aspects of your NVIDIA GPU setup.
+      options:
+        - label: Check that the all CUDA dependencies are installed and are compatible with your GPU (refer to [CUDA's documentation](https://docs.nvidia.com/deploy/cuda-compatibility/#frequently-asked-questions))
+        - label: Ensure an NVIDIA GPU is installed and recognized by the system (run `nvidia-smi` to verify).
+        - label: Ensure proper permissions are set for accessing GPU resources.
+        - label: Docker users - Verify that the NVIDIA Container Toolkit is configured correctly (e.g. run `sudo docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi`)

.github/ISSUE_TEMPLATE/bug_report.md

@@ -1,24 +0,0 @@
----
-name: Bug report
-about: Create a report to help us improve
-title: ''
-labels: bug
-assignees: ''
-
----
-
-Note: if you'd like to *ask a question* or *open a discussion*, head over to the [Discussions](https://github.com/imartinez/privateGPT/discussions) section and post it there.
-
-**Describe the bug and how to reproduce it**
-A clear and concise description of what the bug is and the steps to reproduce the behavior.
-
-**Expected behavior**
-A clear and concise description of what you expected to happen.
-
-**Environment (please complete the following information):**
- - OS / hardware: [e.g. macOS 12.6 / M1]
- - Python version [e.g. 3.11.3]
-- Other relevant information
-
-**Additional context**
-Add any other context about the problem here.

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,8 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Documentation
+    url: https://docs.privategpt.dev
+    about: Please refer to our documentation for more details and guidance.
+  - name: Discord
+    url: https://discord.gg/bK6mRVpErU
+    about: Join our Discord community to ask questions and get help.

.github/ISSUE_TEMPLATE/docs.yml

@@ -0,0 +1,19 @@
+name: Documentation
+description: Suggest a change or addition to the documentation.
+title: "[DOCS] "
+labels: ["documentation"]
+
+body:
+  - type: markdown
+    attributes:
+      value: |
+        **Please describe the documentation change or addition you would like to suggest.**
+
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      description: Provide a detailed description of the documentation change.
+      placeholder: "Detailed description of the documentation change"
+    validations:
+      required: true

.github/ISSUE_TEMPLATE/feature.yml

@@ -0,0 +1,37 @@
+name: Enhancement
+description: Suggest an enhancement or improvement to the project.
+title: "[FEATURE] "
+labels: ["enhancement"]
+
+body:
+  - type: markdown
+    attributes:
+      value: |
+        **Please describe the enhancement or improvement you would like to suggest.**
+
+  - type: textarea
+    id: feature_description
+    attributes:
+      label: Feature Description
+      description: Provide a detailed description of the enhancement.
+      placeholder: "Detailed description of the enhancement"
+    validations:
+      required: true
+
+  - type: textarea
+    id: reason
+    attributes:
+      label: Reason
+      description: Explain the reason for this enhancement.
+      placeholder: "Reason for the enhancement"
+    validations:
+      required: true
+
+  - type: textarea
+    id: value
+    attributes:
+      label: Value of Feature
+      description: Describe the value or benefits this feature will bring.
+      placeholder: "Value or benefits of the feature"
+    validations:
+      required: true

.github/ISSUE_TEMPLATE/feature_request.md

@@ -1,22 +0,0 @@
----
-name: Feature request
-about: Suggest an idea for this project
-title: ''
-labels: enhancement
-assignees: ''
-
----
-
-Note: if you'd like to *ask a question* or *open a discussion*, head over to the [Discussions](https://github.com/imartinez/privateGPT/discussions) section and post it there.
-
-**Is your feature request related to a problem? Please describe.**
-A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
-
-**Describe the solution you'd like**
-A clear and concise description of what you want to happen.
-
-**Describe alternatives you've considered**
-A clear and concise description of any alternative solutions or features you've considered.
-
-**Additional context**
-Add any other context or screenshots about the feature request here.

.github/ISSUE_TEMPLATE/question.yml

@@ -0,0 +1,19 @@
+name: Question
+description: Ask a question about the project.
+title: "[QUESTION] "
+labels: ["question"]
+
+body:
+  - type: markdown
+    attributes:
+      value: |
+        **Please describe your question in detail.**
+
+  - type: textarea
+    id: question
+    attributes:
+      label: Question
+      description: Provide a detailed description of your question.
+      placeholder: "Detailed description of the question"
+    validations:
+      required: true

.github/pull_request_template.md

@@ -0,0 +1,37 @@
+# Description
+
+Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.
+
+## Type of Change
+
+Please delete options that are not relevant.
+
+- [ ] Bug fix (non-breaking change which fixes an issue)
+- [ ] New feature (non-breaking change which adds functionality)
+- [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
+- [ ] This change requires a documentation update
+
+## How Has This Been Tested?
+
+Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration
+
+- [ ] Added new unit/integration tests
+- [ ] I stared at the code and made sure it makes sense
+
+**Test Configuration**:
+* Firmware version:
+* Hardware:
+* Toolchain:
+* SDK:
+
+## Checklist:
+
+- [ ] My code follows the style guidelines of this project
+- [ ] I have performed a self-review of my code
+- [ ] I have commented my code, particularly in hard-to-understand areas
+- [ ] I have made corresponding changes to the documentation
+- [ ] My changes generate no new warnings
+- [ ] I have added tests that prove my fix is effective or that my feature works
+- [ ] New and existing unit tests pass locally with my changes
+- [ ] Any dependent changes have been merged and published in downstream modules
+- [ ] I ran `make check; make test` to ensure mypy and tests pass

.github/workflows/actions/install_dependencies/action.yml

@@ -0,0 +1,30 @@
+name: "Install Dependencies"
+description: "Action to build the project dependencies from the main versions"
+inputs:
+  python_version:
+    required: true
+    type: string
+    default: "3.11.4"
+  poetry_version:
+    required: true
+    type: string
+    default: "1.8.3"
+
+runs:
+  using: composite
+  steps:
+    - name: Install Poetry
+      uses: snok/install-poetry@v1
+      with:
+        version: ${{ inputs.poetry_version }}
+        virtualenvs-create: true
+        virtualenvs-in-project: false
+        installer-parallel: true
+    - uses: actions/setup-python@v4
+      with:
+        python-version: ${{ inputs.python_version }}
+        cache: "poetry"
+    - name: Install Dependencies
+      run: poetry install --extras "ui vector-stores-qdrant" --no-root
+      shell: bash
+

.github/workflows/fern-check.yml

@@ -0,0 +1,21 @@
+name: fern check
+
+on:
+  pull_request:
+    branches:
+      - main
+    paths:
+      - "fern/**"
+      
+jobs:
+  fern-check:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v4
+
+      - name: Install Fern
+        run:  npm install -g fern-api
+
+      - name: Check Fern API is valid
+        run: fern check

.github/workflows/generate-release.yml

@@ -0,0 +1,83 @@
+name: generate-release
+
+on:
+  release:
+    types: [ published ]
+  workflow_dispatch:
+
+env:
+  REGISTRY: docker.io
+  IMAGE_NAME: ${{ github.repository }}
+  platforms: linux/amd64,linux/arm64
+  DEFAULT_TYPE: "ollama"
+
+jobs:
+  build-and-push-image:
+    runs-on: ubuntu-latest
+
+    strategy:
+      matrix:
+        type: [ llamacpp-cpu, ollama ]
+
+    permissions:
+      contents: read
+      packages: write
+
+    outputs:
+      version: ${{ steps.version.outputs.version }}
+
+    steps:
+      - name: Free Disk Space (Ubuntu)
+        uses: jlumbroso/free-disk-space@main
+        with:
+          tool-cache: false
+          android: true
+          dotnet: true
+          haskell: true
+          large-packages: true
+          docker-images: false
+          swap-storage: true
+
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_PASSWORD }}
+
+      - name: Extract metadata (tags, labels) for Docker
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          tags: |
+            type=semver,pattern={{version}},enable=${{ matrix.type == env.DEFAULT_TYPE }}
+            type=semver,pattern={{version}}-${{ matrix.type }}
+            type=semver,pattern={{major}}.{{minor}},enable=${{ matrix.type == env.DEFAULT_TYPE }}
+            type=semver,pattern={{major}}.{{minor}}-${{ matrix.type }}
+            type=raw,value=latest,enable=${{ matrix.type == env.DEFAULT_TYPE }}
+            type=sha
+          flavor: |
+            latest=false
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          file: Dockerfile.${{ matrix.type }}
+          platforms: ${{ env.platforms }}
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+
+      - name: Version output
+        id: version
+        run: echo "version=${{ steps.meta.outputs.version }}" >> "$GITHUB_OUTPUT"

.github/workflows/preview-docs.yml

@@ -0,0 +1,54 @@
+name: deploy preview docs
+
+on:
+  pull_request_target:
+    branches:
+      - main
+    paths:
+      - "fern/**"
+
+jobs:
+  preview-docs:
+    runs-on: ubuntu-latest
+
+    permissions:
+      contents: read
+      pull-requests: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          ref: refs/pull/${{ github.event.pull_request.number }}/merge
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: "18"
+
+      - name: Install Fern
+        run: npm install -g fern-api
+
+      - name: Generate Documentation Preview with Fern
+        id: generate_docs
+        env:
+          FERN_TOKEN: ${{ secrets.FERN_TOKEN }}
+        run: |
+          output=$(fern generate --docs --preview --log-level debug)
+          echo "$output"
+          # Extract the URL
+          preview_url=$(echo "$output" | grep -oP '(?<=Published docs to )https://[^\s]*')
+          # Set the output for the step
+          echo "::set-output name=preview_url::$preview_url"
+      - name: Comment PR with URL using github-actions bot
+        uses: actions/github-script@v7
+        if: ${{ steps.generate_docs.outputs.preview_url }}
+        with:
+          script: |
+            const preview_url = '${{ steps.generate_docs.outputs.preview_url }}';
+            github.rest.issues.createComment({
+              issue_number: context.issue.number,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body: `Published docs preview URL: ${preview_url}`
+            })

.github/workflows/publish-docs.yml

@@ -0,0 +1,26 @@
+name: publish docs
+
+on: 
+  push: 
+    branches: 
+      - main
+    paths:
+      - "fern/**"
+
+jobs:
+  publish-docs:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v4
+
+      - name: Setup node
+        uses: actions/setup-node@v3
+
+      - name: Download Fern
+        run: npm install -g fern-api
+
+      - name: Generate and Publish Docs
+        env:
+          FERN_TOKEN: ${{ secrets.FERN_TOKEN }}
+        run: fern generate --docs --log-level debug

.github/workflows/release-please.yml

@@ -0,0 +1,19 @@
+name: release-please
+
+on:
+  push:
+    branches:
+      - main
+
+permissions:
+  contents: write
+  pull-requests: write
+
+jobs:
+  release-please:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: google-github-actions/release-please-action@v3
+        with:
+          release-type: simple
+          version-file: version.txt

.github/workflows/stale.yml

@@ -0,0 +1,30 @@
+# This workflow warns and then closes issues and PRs that have had no activity for a specified amount of time.
+#
+# You can adjust the behavior by modifying this file.
+# For more information, see:
+# https://github.com/actions/stale
+name: Mark stale issues and pull requests
+
+on:
+  schedule:
+  - cron: '42 5 * * *'
+
+jobs:
+  stale:
+
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+
+    steps:
+    - uses: actions/stale@v8
+      with:
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+        days-before-stale: 15
+        stale-issue-message: 'Stale issue'
+        stale-pr-message: 'Stale pull request'
+        stale-issue-label: 'stale'
+        stale-pr-label: 'stale'
+        exempt-issue-labels: 'autorelease: pending'
+        exempt-pr-labels: 'autorelease: pending'

.github/workflows/tests.yml

@@ -0,0 +1,67 @@
+name: tests
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.head_ref || github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}
+
+jobs:
+  setup:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: ./.github/workflows/actions/install_dependencies
+
+  checks:
+    needs: setup
+    runs-on: ubuntu-latest
+    name: ${{ matrix.quality-command }}
+    strategy:
+      matrix:
+        quality-command:
+          - black
+          - ruff
+          - mypy
+    steps:
+      - uses: actions/checkout@v3
+      - uses: ./.github/workflows/actions/install_dependencies
+      - name: run ${{ matrix.quality-command }}
+        run: make ${{ matrix.quality-command }}
+
+  test:
+    needs: setup
+    runs-on: ubuntu-latest
+    name: test
+    steps:
+      - uses: actions/checkout@v3
+      - uses: ./.github/workflows/actions/install_dependencies
+      - name: run test
+        run: make test-coverage
+      # Run even if make test fails for coverage reports
+      # TODO: select a better xml results displayer
+      - name: Archive test results coverage results
+        uses: actions/upload-artifact@v3
+        if: always()
+        with:
+          name: test_results
+          path: tests-results.xml
+      - name: Archive code coverage results
+        uses: actions/upload-artifact@v3
+        if: always()
+        with:
+          name: code-coverage-report
+          path: htmlcov/
+
+  all_checks_passed:
+    # Used to easily force requirements checks in GitHub
+    needs:
+      - checks
+      - test
+    runs-on: ubuntu-latest
+    steps:
+      - run: echo "All checks passed"

.gitignore

@@ -1,174 +1,31 @@
-# OSX
-.DS_STORE
+.venv
+.env
+venv
 
-# Models
-models/
+settings-me.yaml
 
-# Local Chroma db
-.chroma/
-db/
-persist_directory/chroma.sqlite
+.ruff_cache
+.pytest_cache
+.mypy_cache
 
-# Byte-compiled / optimized / DLL files
+# byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
-*$py.class
 
-# C extensions
-*.so
-
-# Distribution / packaging
-.Python
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-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
+# unit tests / coverage reports
+/tests-results.xml
+/.coverage
+/coverage.xml
+/htmlcov/
 
 # 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
+/.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
+# IDE
+.idea/
+.vscode/
+/.run/
+.fleet/
 
-# 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/
-
-# vscode
-.vscode/launch.json
+# macOS
+.DS_Store

.pre-commit-config.yaml

@@ -1,44 +1,43 @@
----
-files: ^(.*\.(py|json|md|sh|yaml|cfg|txt))$
-exclude: ^(\.[^/]*cache/.*|.*/_user.py|source_documents/)$
+default_install_hook_types:
+# Mandatory to install both pre-commit and pre-push hooks (see https://pre-commit.com/#top_level-default_install_hook_types)
+# Add new hook types here to ensure automatic installation when running `pre-commit install`
+- pre-commit
+- pre-push
 repos:
-  - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
-    hooks:
-      #- id: no-commit-to-branch
-      #  args: [--branch, main]
-      - id: check-yaml
-        args: [--unsafe]
-      # - id: debug-statements
-      - id: end-of-file-fixer
-      - id: trailing-whitespace
-        exclude-files: \.md$
-      - id: check-json
-      - id: mixed-line-ending
-      # - id: check-builtin-literals
-      # - id: check-ast
-      - id: check-merge-conflict
-      - id: check-executables-have-shebangs
-      - id: check-shebang-scripts-are-executable
-      - id: check-docstring-first
-      - id: fix-byte-order-marker
-      - id: check-case-conflict
-      # - id: check-toml
-  - repo: https://github.com/adrienverge/yamllint.git
-    rev: v1.29.0
-    hooks:
-      - id: yamllint
-        args:
-          - --no-warnings
-          - -d
-          - '{extends: relaxed, rules: {line-length: {max: 90}}}'
-  - repo: https://github.com/codespell-project/codespell
-    rev: v2.2.2
-    hooks:
-      - id: codespell
-        args:
-          # - --builtin=clear,rare,informal,usage,code,names,en-GB_to_en-US
-          - --builtin=clear,rare,informal,usage,code,names
-          - --ignore-words-list=hass,master
-          - --skip="./.*"
-          - --quiet-level=2
+- repo: https://github.com/pre-commit/pre-commit-hooks
+  rev: v4.3.0
+  hooks:
+  - id: trailing-whitespace
+  - id: end-of-file-fixer
+  - id: check-yaml
+  - id: check-json
+  - id: check-added-large-files
+
+- repo: local
+  hooks:
+  - id: black
+    name: Formatting (black)
+    entry: black
+    language: system
+    types: [python]
+    stages: [commit]
+  - id: ruff
+    name: Linter (ruff)
+    entry: ruff
+    language: system
+    types: [python]
+    stages: [commit]
+  - id: mypy
+    name: Type checking (mypy)
+    entry: make mypy
+    pass_filenames: false
+    language: system
+    types: [python]
+    stages: [commit]
+  - id: test
+    name: Unit tests (pytest)
+    entry: make test
+    pass_filenames: false
+    language: system
+    types: [python]
+    stages: [push]

CHANGELOG.md

@@ -0,0 +1,163 @@
+# Changelog
+
+## [0.6.1](https://github.com/zylon-ai/private-gpt/compare/v0.6.0...v0.6.1) (2024-08-05)
+
+
+### Bug Fixes
+
+* add built image from DockerHub ([#2042](https://github.com/zylon-ai/private-gpt/issues/2042)) ([f09f6dd](https://github.com/zylon-ai/private-gpt/commit/f09f6dd2553077d4566dbe6b48a450e05c2f049e))
+* Adding azopenai to model list ([#2035](https://github.com/zylon-ai/private-gpt/issues/2035)) ([1c665f7](https://github.com/zylon-ai/private-gpt/commit/1c665f7900658144f62814b51f6e3434a6d7377f))
+* **deploy:** generate docker release when new version is released ([#2038](https://github.com/zylon-ai/private-gpt/issues/2038)) ([1d4c14d](https://github.com/zylon-ai/private-gpt/commit/1d4c14d7a3c383c874b323d934be01afbaca899e))
+* **deploy:** improve Docker-Compose and quickstart on Docker ([#2037](https://github.com/zylon-ai/private-gpt/issues/2037)) ([dae0727](https://github.com/zylon-ai/private-gpt/commit/dae0727a1b4abd35d2b0851fe30e0a4ed67e0fbb))
+
+## [0.6.0](https://github.com/zylon-ai/private-gpt/compare/v0.5.0...v0.6.0) (2024-08-02)
+
+
+### Features
+
+* bump dependencies ([#1987](https://github.com/zylon-ai/private-gpt/issues/1987)) ([b687dc8](https://github.com/zylon-ai/private-gpt/commit/b687dc852413404c52d26dcb94536351a63b169d))
+* **docs:** add privategpt-ts sdk ([#1924](https://github.com/zylon-ai/private-gpt/issues/1924)) ([d13029a](https://github.com/zylon-ai/private-gpt/commit/d13029a046f6e19e8ee65bef3acd96365c738df2))
+* **docs:** Fix setup docu ([#1926](https://github.com/zylon-ai/private-gpt/issues/1926)) ([067a5f1](https://github.com/zylon-ai/private-gpt/commit/067a5f144ca6e605c99d7dbe9ca7d8207ac8808d))
+* **docs:** update doc for ipex-llm ([#1968](https://github.com/zylon-ai/private-gpt/issues/1968)) ([19a7c06](https://github.com/zylon-ai/private-gpt/commit/19a7c065ef7f42b37f289dd28ac945f7afc0e73a))
+* **docs:** update documentation and fix preview-docs ([#2000](https://github.com/zylon-ai/private-gpt/issues/2000)) ([4523a30](https://github.com/zylon-ai/private-gpt/commit/4523a30c8f004aac7a7ae224671e2c45ec0cb973))
+* **llm:** add progress bar when ollama is pulling models ([#2031](https://github.com/zylon-ai/private-gpt/issues/2031)) ([cf61bf7](https://github.com/zylon-ai/private-gpt/commit/cf61bf780f8d122e4057d002abf03563bb45614a))
+* **llm:** autopull ollama models ([#2019](https://github.com/zylon-ai/private-gpt/issues/2019)) ([20bad17](https://github.com/zylon-ai/private-gpt/commit/20bad17c9857809158e689e9671402136c1e3d84))
+* **llm:** Support for Google Gemini LLMs and Embeddings ([#1965](https://github.com/zylon-ai/private-gpt/issues/1965)) ([fc13368](https://github.com/zylon-ai/private-gpt/commit/fc13368bc72d1f4c27644677431420ed77731c03))
+* make llama3.1 as default ([#2022](https://github.com/zylon-ai/private-gpt/issues/2022)) ([9027d69](https://github.com/zylon-ai/private-gpt/commit/9027d695c11fbb01e62424b855665de71d513417))
+* prompt_style applied to all LLMs + extra LLM params. ([#1835](https://github.com/zylon-ai/private-gpt/issues/1835)) ([e21bf20](https://github.com/zylon-ai/private-gpt/commit/e21bf20c10938b24711d9f2c765997f44d7e02a9))
+* **recipe:** add our first recipe  `Summarize` ([#2028](https://github.com/zylon-ai/private-gpt/issues/2028)) ([8119842](https://github.com/zylon-ai/private-gpt/commit/8119842ae6f1f5ecfaf42b06fa0d1ffec675def4))
+* **vectordb:** Milvus vector db Integration ([#1996](https://github.com/zylon-ai/private-gpt/issues/1996)) ([43cc31f](https://github.com/zylon-ai/private-gpt/commit/43cc31f74015f8d8fcbf7a8ea7d7d9ecc66cf8c9))
+* **vectorstore:** Add clickhouse support as vectore store ([#1883](https://github.com/zylon-ai/private-gpt/issues/1883)) ([2612928](https://github.com/zylon-ai/private-gpt/commit/26129288394c7483e6fc0496a11dc35679528cc1))
+
+
+### Bug Fixes
+
+* "no such group" error in Dockerfile, added docx2txt and cryptography deps ([#1841](https://github.com/zylon-ai/private-gpt/issues/1841)) ([947e737](https://github.com/zylon-ai/private-gpt/commit/947e737f300adf621d2261d527192f36f3387f8e))
+* **config:** make tokenizer optional and include a troubleshooting doc ([#1998](https://github.com/zylon-ai/private-gpt/issues/1998)) ([01b7ccd](https://github.com/zylon-ai/private-gpt/commit/01b7ccd0648be032846647c9a184925d3682f612))
+* **docs:** Fix concepts.mdx referencing to installation page ([#1779](https://github.com/zylon-ai/private-gpt/issues/1779)) ([dde0224](https://github.com/zylon-ai/private-gpt/commit/dde02245bcd51a7ede7b6789c82ae217cac53d92))
+* **docs:** Update installation.mdx ([#1866](https://github.com/zylon-ai/private-gpt/issues/1866)) ([c1802e7](https://github.com/zylon-ai/private-gpt/commit/c1802e7cf0e56a2603213ec3b6a4af8fadb8a17a))
+* ffmpy dependency ([#2020](https://github.com/zylon-ai/private-gpt/issues/2020)) ([dabf556](https://github.com/zylon-ai/private-gpt/commit/dabf556dae9cb00fe0262270e5138d982585682e))
+* light mode ([#2025](https://github.com/zylon-ai/private-gpt/issues/2025)) ([1020cd5](https://github.com/zylon-ai/private-gpt/commit/1020cd53288af71a17882781f392512568f1b846))
+* **LLM:** mistral ignoring assistant messages ([#1954](https://github.com/zylon-ai/private-gpt/issues/1954)) ([c7212ac](https://github.com/zylon-ai/private-gpt/commit/c7212ac7cc891f9e3c713cc206ae9807c5dfdeb6))
+* **llm:** special tokens and leading space ([#1831](https://github.com/zylon-ai/private-gpt/issues/1831)) ([347be64](https://github.com/zylon-ai/private-gpt/commit/347be643f7929c56382a77c3f45f0867605e0e0a))
+* make embedding_api_base match api_base when on docker ([#1859](https://github.com/zylon-ai/private-gpt/issues/1859)) ([2a432bf](https://github.com/zylon-ai/private-gpt/commit/2a432bf9c5582a94eb4052b1e80cabdb118d298e))
+* nomic embeddings ([#2030](https://github.com/zylon-ai/private-gpt/issues/2030)) ([5465958](https://github.com/zylon-ai/private-gpt/commit/54659588b5b109a3dd17cca835e275240464d275))
+* prevent to ingest local files (by default) ([#2010](https://github.com/zylon-ai/private-gpt/issues/2010)) ([e54a8fe](https://github.com/zylon-ai/private-gpt/commit/e54a8fe0433252808d0a60f6a08a43c9f5a42f3b))
+* Replacing unsafe `eval()` with `json.loads()` ([#1890](https://github.com/zylon-ai/private-gpt/issues/1890)) ([9d0d614](https://github.com/zylon-ai/private-gpt/commit/9d0d614706581a8bfa57db45f62f84ab23d26f15))
+* **settings:** enable cors by default so it will work when using ts sdk (spa) ([#1925](https://github.com/zylon-ai/private-gpt/issues/1925)) ([966af47](https://github.com/zylon-ai/private-gpt/commit/966af4771dbe5cf3fdf554b5fdf8f732407859c4))
+* **ui:** gradio bug fixes ([#2021](https://github.com/zylon-ai/private-gpt/issues/2021)) ([d4375d0](https://github.com/zylon-ai/private-gpt/commit/d4375d078f18ba53562fd71651159f997fff865f))
+* unify embedding models ([#2027](https://github.com/zylon-ai/private-gpt/issues/2027)) ([40638a1](https://github.com/zylon-ai/private-gpt/commit/40638a18a5713d60fec8fe52796dcce66d88258c))
+
+## [0.5.0](https://github.com/zylon-ai/private-gpt/compare/v0.4.0...v0.5.0) (2024-04-02)
+
+
+### Features
+
+* **code:** improve concat of strings in ui ([#1785](https://github.com/zylon-ai/private-gpt/issues/1785)) ([bac818a](https://github.com/zylon-ai/private-gpt/commit/bac818add51b104cda925b8f1f7b51448e935ca1))
+* **docker:** set default Docker to use Ollama ([#1812](https://github.com/zylon-ai/private-gpt/issues/1812)) ([f83abff](https://github.com/zylon-ai/private-gpt/commit/f83abff8bc955a6952c92cc7bcb8985fcec93afa))
+* **docs:** Add guide Llama-CPP Linux AMD GPU support ([#1782](https://github.com/zylon-ai/private-gpt/issues/1782)) ([8a836e4](https://github.com/zylon-ai/private-gpt/commit/8a836e4651543f099c59e2bf497ab8c55a7cd2e5))
+* **docs:** Feature/upgrade docs ([#1741](https://github.com/zylon-ai/private-gpt/issues/1741)) ([5725181](https://github.com/zylon-ai/private-gpt/commit/572518143ac46532382db70bed6f73b5082302c1))
+* **docs:** upgrade fern ([#1596](https://github.com/zylon-ai/private-gpt/issues/1596)) ([84ad16a](https://github.com/zylon-ai/private-gpt/commit/84ad16af80191597a953248ce66e963180e8ddec))
+* **ingest:** Created a faster ingestion mode - pipeline ([#1750](https://github.com/zylon-ai/private-gpt/issues/1750)) ([134fc54](https://github.com/zylon-ai/private-gpt/commit/134fc54d7d636be91680dc531f5cbe2c5892ac56))
+* **llm - embed:** Add support for Azure OpenAI ([#1698](https://github.com/zylon-ai/private-gpt/issues/1698)) ([1efac6a](https://github.com/zylon-ai/private-gpt/commit/1efac6a3fe19e4d62325e2c2915cd84ea277f04f))
+* **llm:** adds serveral settings for llamacpp and ollama ([#1703](https://github.com/zylon-ai/private-gpt/issues/1703)) ([02dc83e](https://github.com/zylon-ai/private-gpt/commit/02dc83e8e9f7ada181ff813f25051bbdff7b7c6b))
+* **llm:** Ollama LLM-Embeddings decouple + longer keep_alive settings ([#1800](https://github.com/zylon-ai/private-gpt/issues/1800)) ([b3b0140](https://github.com/zylon-ai/private-gpt/commit/b3b0140e244e7a313bfaf4ef10eb0f7e4192710e))
+* **llm:** Ollama timeout setting ([#1773](https://github.com/zylon-ai/private-gpt/issues/1773)) ([6f6c785](https://github.com/zylon-ai/private-gpt/commit/6f6c785dac2bbad37d0b67fda215784298514d39))
+* **local:** tiktoken cache within repo for offline ([#1467](https://github.com/zylon-ai/private-gpt/issues/1467)) ([821bca3](https://github.com/zylon-ai/private-gpt/commit/821bca32e9ee7c909fd6488445ff6a04463bf91b))
+* **nodestore:** add Postgres for the doc and index store ([#1706](https://github.com/zylon-ai/private-gpt/issues/1706)) ([68b3a34](https://github.com/zylon-ai/private-gpt/commit/68b3a34b032a08ca073a687d2058f926032495b3))
+* **rag:** expose similarity_top_k and similarity_score to settings ([#1771](https://github.com/zylon-ai/private-gpt/issues/1771)) ([087cb0b](https://github.com/zylon-ai/private-gpt/commit/087cb0b7b74c3eb80f4f60b47b3a021c81272ae1))
+* **RAG:** Introduce SentenceTransformer Reranker ([#1810](https://github.com/zylon-ai/private-gpt/issues/1810)) ([83adc12](https://github.com/zylon-ai/private-gpt/commit/83adc12a8ef0fa0c13a0dec084fa596445fc9075))
+* **scripts:** Wipe qdrant and obtain db Stats command ([#1783](https://github.com/zylon-ai/private-gpt/issues/1783)) ([ea153fb](https://github.com/zylon-ai/private-gpt/commit/ea153fb92f1f61f64c0d04fff0048d4d00b6f8d0))
+* **ui:** Add Model Information to ChatInterface label ([f0b174c](https://github.com/zylon-ai/private-gpt/commit/f0b174c097c2d5e52deae8ef88de30a0d9013a38))
+* **ui:** add sources check to not repeat identical sources ([#1705](https://github.com/zylon-ai/private-gpt/issues/1705)) ([290b9fb](https://github.com/zylon-ai/private-gpt/commit/290b9fb084632216300e89bdadbfeb0380724b12))
+* **UI:** Faster startup and document listing ([#1763](https://github.com/zylon-ai/private-gpt/issues/1763)) ([348df78](https://github.com/zylon-ai/private-gpt/commit/348df781b51606b2f9810bcd46f850e54192fd16))
+* **ui:** maintain score order when curating sources ([#1643](https://github.com/zylon-ai/private-gpt/issues/1643)) ([410bf7a](https://github.com/zylon-ai/private-gpt/commit/410bf7a71f17e77c4aec723ab80c233b53765964))
+* unify settings for vector and nodestore connections to PostgreSQL ([#1730](https://github.com/zylon-ai/private-gpt/issues/1730)) ([63de7e4](https://github.com/zylon-ai/private-gpt/commit/63de7e4930ac90dd87620225112a22ffcbbb31ee))
+* wipe per storage type ([#1772](https://github.com/zylon-ai/private-gpt/issues/1772)) ([c2d6948](https://github.com/zylon-ai/private-gpt/commit/c2d694852b4696834962a42fde047b728722ad74))
+
+
+### Bug Fixes
+
+* **docs:** Minor documentation amendment ([#1739](https://github.com/zylon-ai/private-gpt/issues/1739)) ([258d02d](https://github.com/zylon-ai/private-gpt/commit/258d02d87c5cb81d6c3a6f06aa69339b670dffa9))
+* Fixed docker-compose ([#1758](https://github.com/zylon-ai/private-gpt/issues/1758)) ([774e256](https://github.com/zylon-ai/private-gpt/commit/774e2560520dc31146561d09a2eb464c68593871))
+* **ingest:** update script label ([#1770](https://github.com/zylon-ai/private-gpt/issues/1770)) ([7d2de5c](https://github.com/zylon-ai/private-gpt/commit/7d2de5c96fd42e339b26269b3155791311ef1d08))
+* **settings:** set default tokenizer to avoid running make setup fail ([#1709](https://github.com/zylon-ai/private-gpt/issues/1709)) ([d17c34e](https://github.com/zylon-ai/private-gpt/commit/d17c34e81a84518086b93605b15032e2482377f7))
+
+## [0.4.0](https://github.com/imartinez/privateGPT/compare/v0.3.0...v0.4.0) (2024-03-06)
+
+
+### Features
+
+* Upgrade to LlamaIndex to 0.10 ([#1663](https://github.com/imartinez/privateGPT/issues/1663)) ([45f0571](https://github.com/imartinez/privateGPT/commit/45f05711eb71ffccdedb26f37e680ced55795d44))
+* **Vector:** support pgvector ([#1624](https://github.com/imartinez/privateGPT/issues/1624)) ([cd40e39](https://github.com/imartinez/privateGPT/commit/cd40e3982b780b548b9eea6438c759f1c22743a8))
+
+## [0.3.0](https://github.com/imartinez/privateGPT/compare/v0.2.0...v0.3.0) (2024-02-16)
+
+
+### Features
+
+* add mistral + chatml prompts ([#1426](https://github.com/imartinez/privateGPT/issues/1426)) ([e326126](https://github.com/imartinez/privateGPT/commit/e326126d0d4cd7e46a79f080c442c86f6dd4d24b))
+* Add stream information to generate SDKs ([#1569](https://github.com/imartinez/privateGPT/issues/1569)) ([24fae66](https://github.com/imartinez/privateGPT/commit/24fae660e6913aac6b52745fb2c2fe128ba2eb79))
+* **API:** Ingest plain text ([#1417](https://github.com/imartinez/privateGPT/issues/1417)) ([6eeb95e](https://github.com/imartinez/privateGPT/commit/6eeb95ec7f17a618aaa47f5034ee5bccae02b667))
+* **bulk-ingest:** Add --ignored Flag to Exclude Specific Files and Directories During Ingestion ([#1432](https://github.com/imartinez/privateGPT/issues/1432)) ([b178b51](https://github.com/imartinez/privateGPT/commit/b178b514519550e355baf0f4f3f6beb73dca7df2))
+* **llm:** Add openailike llm mode ([#1447](https://github.com/imartinez/privateGPT/issues/1447)) ([2d27a9f](https://github.com/imartinez/privateGPT/commit/2d27a9f956d672cb1fe715cf0acdd35c37f378a5)), closes [#1424](https://github.com/imartinez/privateGPT/issues/1424)
+* **llm:** Add support for Ollama LLM ([#1526](https://github.com/imartinez/privateGPT/issues/1526)) ([6bbec79](https://github.com/imartinez/privateGPT/commit/6bbec79583b7f28d9bea4b39c099ebef149db843))
+* **settings:** Configurable context_window and tokenizer ([#1437](https://github.com/imartinez/privateGPT/issues/1437)) ([4780540](https://github.com/imartinez/privateGPT/commit/47805408703c23f0fd5cab52338142c1886b450b))
+* **settings:** Update default model to TheBloke/Mistral-7B-Instruct-v0.2-GGUF ([#1415](https://github.com/imartinez/privateGPT/issues/1415)) ([8ec7cf4](https://github.com/imartinez/privateGPT/commit/8ec7cf49f40701a4f2156c48eb2fad9fe6220629))
+* **ui:** make chat area stretch to fill the screen ([#1397](https://github.com/imartinez/privateGPT/issues/1397)) ([c71ae7c](https://github.com/imartinez/privateGPT/commit/c71ae7cee92463bbc5ea9c434eab9f99166e1363))
+* **UI:** Select file to Query or Delete + Delete ALL ([#1612](https://github.com/imartinez/privateGPT/issues/1612)) ([aa13afd](https://github.com/imartinez/privateGPT/commit/aa13afde07122f2ddda3942f630e5cadc7e4e1ee))
+
+
+### Bug Fixes
+
+* Adding an LLM param to fix broken generator from llamacpp ([#1519](https://github.com/imartinez/privateGPT/issues/1519)) ([869233f](https://github.com/imartinez/privateGPT/commit/869233f0e4f03dc23e5fae43cf7cb55350afdee9))
+* **deploy:** fix local and external dockerfiles ([fde2b94](https://github.com/imartinez/privateGPT/commit/fde2b942bc03688701ed563be6d7d597c75e4e4e))
+* **docker:** docker broken copy ([#1419](https://github.com/imartinez/privateGPT/issues/1419)) ([059f358](https://github.com/imartinez/privateGPT/commit/059f35840adbc3fb93d847d6decf6da32d08670c))
+* **docs:** Update quickstart doc and set version in pyproject.toml to 0.2.0 ([0a89d76](https://github.com/imartinez/privateGPT/commit/0a89d76cc5ed4371ffe8068858f23dfbb5e8cc37))
+* minor bug in chat stream output - python error being serialized ([#1449](https://github.com/imartinez/privateGPT/issues/1449)) ([6191bcd](https://github.com/imartinez/privateGPT/commit/6191bcdbd6e92b6f4d5995967dc196c9348c5954))
+* **settings:** correct yaml multiline string ([#1403](https://github.com/imartinez/privateGPT/issues/1403)) ([2564f8d](https://github.com/imartinez/privateGPT/commit/2564f8d2bb8c4332a6a0ab6d722a2ac15006b85f))
+* **tests:** load the test settings only when running tests ([d3acd85](https://github.com/imartinez/privateGPT/commit/d3acd85fe34030f8cfd7daf50b30c534087bdf2b))
+* **UI:** Updated ui.py. Frees up the CPU to not be bottlenecked. ([24fb80c](https://github.com/imartinez/privateGPT/commit/24fb80ca38f21910fe4fd81505d14960e9ed4faa))
+
+## [0.2.0](https://github.com/imartinez/privateGPT/compare/v0.1.0...v0.2.0) (2023-12-10)
+
+
+### Features
+
+* **llm:** drop default_system_prompt ([#1385](https://github.com/imartinez/privateGPT/issues/1385)) ([a3ed14c](https://github.com/imartinez/privateGPT/commit/a3ed14c58f77351dbd5f8f2d7868d1642a44f017))
+* **ui:** Allows User to Set System Prompt via "Additional Options" in Chat Interface ([#1353](https://github.com/imartinez/privateGPT/issues/1353)) ([145f3ec](https://github.com/imartinez/privateGPT/commit/145f3ec9f41c4def5abf4065a06fb0786e2d992a))
+
+## [0.1.0](https://github.com/imartinez/privateGPT/compare/v0.0.2...v0.1.0) (2023-11-30)
+
+
+### Features
+
+* Disable Gradio Analytics ([#1165](https://github.com/imartinez/privateGPT/issues/1165)) ([6583dc8](https://github.com/imartinez/privateGPT/commit/6583dc84c082773443fc3973b1cdf8095fa3fec3))
+* Drop loguru and use builtin `logging` ([#1133](https://github.com/imartinez/privateGPT/issues/1133)) ([64c5ae2](https://github.com/imartinez/privateGPT/commit/64c5ae214a9520151c9c2d52ece535867d799367))
+* enable resume download for hf_hub_download ([#1249](https://github.com/imartinez/privateGPT/issues/1249)) ([4197ada](https://github.com/imartinez/privateGPT/commit/4197ada6267c822f32c1d7ba2be6e7ce145a3404))
+* move torch and transformers to local group ([#1172](https://github.com/imartinez/privateGPT/issues/1172)) ([0d677e1](https://github.com/imartinez/privateGPT/commit/0d677e10b970aec222ec04837d0f08f1631b6d4a))
+* Qdrant support ([#1228](https://github.com/imartinez/privateGPT/issues/1228)) ([03d1ae6](https://github.com/imartinez/privateGPT/commit/03d1ae6d70dffdd2411f0d4e92f65080fff5a6e2))
+
+
+### Bug Fixes
+
+* Docker and sagemaker setup ([#1118](https://github.com/imartinez/privateGPT/issues/1118)) ([895588b](https://github.com/imartinez/privateGPT/commit/895588b82a06c2bc71a9e22fb840c7f6442a3b5b))
+* fix pytorch version to avoid wheel bug ([#1123](https://github.com/imartinez/privateGPT/issues/1123)) ([24cfddd](https://github.com/imartinez/privateGPT/commit/24cfddd60f74aadd2dade4c63f6012a2489938a1))
+* Remove global state ([#1216](https://github.com/imartinez/privateGPT/issues/1216)) ([022bd71](https://github.com/imartinez/privateGPT/commit/022bd718e3dfc197027b1e24fb97e5525b186db4))
+* sagemaker config and chat methods ([#1142](https://github.com/imartinez/privateGPT/issues/1142)) ([a517a58](https://github.com/imartinez/privateGPT/commit/a517a588c4927aa5c5c2a93e4f82a58f0599d251))
+* typo in README.md ([#1091](https://github.com/imartinez/privateGPT/issues/1091)) ([ba23443](https://github.com/imartinez/privateGPT/commit/ba23443a70d323cd4f9a242b33fd9dce1bacd2db))
+* Windows 11 failing to auto-delete tmp file ([#1260](https://github.com/imartinez/privateGPT/issues/1260)) ([0d52002](https://github.com/imartinez/privateGPT/commit/0d520026a3d5b08a9b8487be992d3095b21e710c))
+* Windows permission error on ingest service tmp files ([#1280](https://github.com/imartinez/privateGPT/issues/1280)) ([f1cbff0](https://github.com/imartinez/privateGPT/commit/f1cbff0fb7059432d9e71473cbdd039032dab60d))
+
+## [0.0.2](https://github.com/imartinez/privateGPT/compare/v0.0.1...v0.0.2) (2023-10-20)
+
+
+### Bug Fixes
+
+* chromadb max batch size ([#1087](https://github.com/imartinez/privateGPT/issues/1087)) ([f5a9bf4](https://github.com/imartinez/privateGPT/commit/f5a9bf4e374b2d4c76438cf8a97cccf222ec8e6f))
+
+## 0.0.1 (2023-10-20)
+
+### Miscellaneous Chores
+
+* Initial version ([490d93f](https://github.com/imartinez/privateGPT/commit/490d93fdc1977443c92f6c42e57a1c585aa59430))

CITATION.cff

@@ -0,0 +1,16 @@
+# This CITATION.cff file was generated with cffinit.
+# Visit https://bit.ly/cffinit to generate yours today!
+
+cff-version: 1.2.0
+title: PrivateGPT
+message: >-
+  If you use this software, please cite it using the
+  metadata from this file.
+type: software
+authors:
+  - name: Zylon by PrivateGPT
+    address: hello@zylon.ai
+    website: 'https://www.zylon.ai/'
+repository-code: 'https://github.com/zylon-ai/private-gpt'
+license: Apache-2.0
+date-released: '2023-05-02'

Dockerfile.llamacpp-cpu

@@ -0,0 +1,62 @@
+### IMPORTANT, THIS IMAGE CAN ONLY BE RUN IN LINUX DOCKER
+### You will run into a segfault in mac
+FROM python:3.11.6-slim-bookworm as base
+
+# Install poetry
+RUN pip install pipx
+RUN python3 -m pipx ensurepath
+RUN pipx install poetry==1.8.3
+ENV PATH="/root/.local/bin:$PATH"
+ENV PATH=".venv/bin/:$PATH"
+
+# Dependencies to build llama-cpp
+RUN apt update && apt install -y \
+  libopenblas-dev\
+  ninja-build\
+  build-essential\
+  pkg-config\
+  wget
+
+# https://python-poetry.org/docs/configuration/#virtualenvsin-project
+ENV POETRY_VIRTUALENVS_IN_PROJECT=true
+
+FROM base as dependencies
+WORKDIR /home/worker/app
+COPY pyproject.toml poetry.lock ./
+
+ARG POETRY_EXTRAS="ui embeddings-huggingface llms-llama-cpp vector-stores-qdrant"
+RUN poetry install --no-root --extras "${POETRY_EXTRAS}"
+
+FROM base as app
+
+ENV PYTHONUNBUFFERED=1
+ENV PORT=8080
+ENV APP_ENV=prod
+ENV PYTHONPATH="$PYTHONPATH:/home/worker/app/private_gpt/"
+EXPOSE 8080
+
+# Prepare a non-root user
+# More info about how to configure UIDs and GIDs in Docker:
+# https://github.com/systemd/systemd/blob/main/docs/UIDS-GIDS.md
+
+# Define the User ID (UID) for the non-root user
+# UID 100 is chosen to avoid conflicts with existing system users
+ARG UID=100
+
+# Define the Group ID (GID) for the non-root user
+# GID 65534 is often used for the 'nogroup' or 'nobody' group
+ARG GID=65534
+
+RUN adduser --system --gid ${GID} --uid ${UID} --home /home/worker worker
+WORKDIR /home/worker/app
+
+RUN chown worker /home/worker/app
+RUN mkdir local_data && chown worker local_data
+RUN mkdir models && chown worker models
+COPY --chown=worker --from=dependencies /home/worker/app/.venv/ .venv
+COPY --chown=worker private_gpt/ private_gpt
+COPY --chown=worker *.yaml ./
+COPY --chown=worker scripts/ scripts
+
+USER worker
+ENTRYPOINT python -m private_gpt

Dockerfile.ollama

@@ -0,0 +1,51 @@
+FROM python:3.11.6-slim-bookworm as base
+
+# Install poetry
+RUN pip install pipx
+RUN python3 -m pipx ensurepath
+RUN pipx install poetry==1.8.3
+ENV PATH="/root/.local/bin:$PATH"
+ENV PATH=".venv/bin/:$PATH"
+
+# https://python-poetry.org/docs/configuration/#virtualenvsin-project
+ENV POETRY_VIRTUALENVS_IN_PROJECT=true
+
+FROM base as dependencies
+WORKDIR /home/worker/app
+COPY pyproject.toml poetry.lock ./
+
+ARG POETRY_EXTRAS="ui vector-stores-qdrant llms-ollama embeddings-ollama"
+RUN poetry install --no-root --extras "${POETRY_EXTRAS}"
+
+FROM base as app
+ENV PYTHONUNBUFFERED=1
+ENV PORT=8080
+ENV APP_ENV=prod
+ENV PYTHONPATH="$PYTHONPATH:/home/worker/app/private_gpt/"
+EXPOSE 8080
+
+# Prepare a non-root user
+# More info about how to configure UIDs and GIDs in Docker:
+# https://github.com/systemd/systemd/blob/main/docs/UIDS-GIDS.md
+
+# Define the User ID (UID) for the non-root user
+# UID 100 is chosen to avoid conflicts with existing system users
+ARG UID=100
+
+# Define the Group ID (GID) for the non-root user
+# GID 65534 is often used for the 'nogroup' or 'nobody' group
+ARG GID=65534
+
+RUN adduser --system --gid ${GID} --uid ${UID} --home /home/worker worker
+WORKDIR /home/worker/app
+
+RUN chown worker /home/worker/app
+RUN mkdir local_data && chown worker local_data
+RUN mkdir models && chown worker models
+COPY --chown=worker --from=dependencies /home/worker/app/.venv/ .venv
+COPY --chown=worker private_gpt/ private_gpt
+COPY --chown=worker *.yaml .
+COPY --chown=worker scripts/ scripts
+
+USER worker
+ENTRYPOINT python -m private_gpt

Makefile

@@ -0,0 +1,78 @@
+# Any args passed to the make script, use with $(call args, default_value)
+args = `arg="$(filter-out $@,$(MAKECMDGOALS))" && echo $${arg:-${1}}`
+
+########################################################################################################################
+# Quality checks
+########################################################################################################################
+
+test:
+	PYTHONPATH=. poetry run pytest tests
+
+test-coverage:
+	PYTHONPATH=. poetry run pytest tests --cov private_gpt --cov-report term --cov-report=html --cov-report xml --junit-xml=tests-results.xml
+
+black:
+	poetry run black . --check
+
+ruff:
+	poetry run ruff check private_gpt tests
+
+format:
+	poetry run black .
+	poetry run ruff check private_gpt tests --fix
+
+mypy:
+	poetry run mypy private_gpt
+
+check:
+	make format
+	make mypy
+
+########################################################################################################################
+# Run
+########################################################################################################################
+
+run:
+	poetry run python -m private_gpt
+
+dev-windows:
+	(set PGPT_PROFILES=local & poetry run python -m uvicorn private_gpt.main:app --reload --port 8001)
+
+dev:
+	PYTHONUNBUFFERED=1 PGPT_PROFILES=local poetry run python -m uvicorn private_gpt.main:app --reload --port 8001
+
+########################################################################################################################
+# Misc
+########################################################################################################################
+
+api-docs:
+	PGPT_PROFILES=mock poetry run python scripts/extract_openapi.py private_gpt.main:app --out fern/openapi/openapi.json
+
+ingest:
+	@poetry run python scripts/ingest_folder.py $(call args)
+
+stats:
+	poetry run python scripts/utils.py stats
+
+wipe:
+	poetry run python scripts/utils.py wipe
+
+setup:
+	poetry run python scripts/setup
+
+list:
+	@echo "Available commands:"
+	@echo "  test            : Run tests using pytest"
+	@echo "  test-coverage   : Run tests with coverage report"
+	@echo "  black           : Check code format with black"
+	@echo "  ruff            : Check code with ruff"
+	@echo "  format          : Format code with black and ruff"
+	@echo "  mypy            : Run mypy for type checking"
+	@echo "  check           : Run format and mypy commands"
+	@echo "  run             : Run the application"
+	@echo "  dev-windows     : Run the application in development mode on Windows"
+	@echo "  dev             : Run the application in development mode"
+	@echo "  api-docs        : Generate API documentation"
+	@echo "  ingest          : Ingest data using specified script"
+	@echo "  wipe            : Wipe data using specified script"
+	@echo "  setup           : Setup the application"

README.md

@@ -1,152 +1,158 @@
-# privateGPT
-Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask questions without an internet connection!
+# 🔒 PrivateGPT 📑
 
-> :ear: **Need help applying PrivateGPT to your specific use case?** [Let us know more about it](https://forms.gle/4cSDmH13RZBHV9at7) and we'll try to help! We are refining PrivateGPT through your feedback.
+[![Tests](https://github.com/zylon-ai/private-gpt/actions/workflows/tests.yml/badge.svg)](https://github.com/zylon-ai/private-gpt/actions/workflows/tests.yml?query=branch%3Amain)
+[![Website](https://img.shields.io/website?up_message=check%20it&down_message=down&url=https%3A%2F%2Fdocs.privategpt.dev%2F&label=Documentation)](https://docs.privategpt.dev/)
+[![Discord](https://img.shields.io/discord/1164200432894234644?logo=discord&label=PrivateGPT)](https://discord.gg/bK6mRVpErU)
+[![X (formerly Twitter) Follow](https://img.shields.io/twitter/follow/ZylonPrivateGPT)](https://twitter.com/ZylonPrivateGPT)
 
-<img width="902" alt="demo" src="https://user-images.githubusercontent.com/721666/236942256-985801c9-25b9-48ef-80be-3acbb4575164.png">
+![Gradio UI](/fern/docs/assets/ui.png?raw=true)
 
-Built with [LangChain](https://github.com/hwchase17/langchain), [LlamaIndex](https://www.llamaindex.ai/), [GPT4All](https://github.com/nomic-ai/gpt4all), [LlamaCpp](https://github.com/ggerganov/llama.cpp), [Chroma](https://www.trychroma.com/) and [SentenceTransformers](https://www.sbert.net/).
+PrivateGPT is a production-ready AI project that allows you to ask questions about your documents using the power
+of Large Language Models (LLMs), even in scenarios without an Internet connection. 100% private, no data leaves your
+execution environment at any point.
 
-# Environment Setup
-In order to set your environment up to run the code here, first install all requirements:
+>[!TIP]
+> If you are looking for an **enterprise-ready, fully private AI workspace**
+> check out [Zylon's website](https://zylon.ai)  or [request a demo](https://cal.com/zylon/demo?source=pgpt-readme).
+> Crafted by the team behind PrivateGPT, Zylon is a best-in-class AI collaborative
+> workspace that can be easily deployed on-premise (data center, bare metal...) or in your private cloud (AWS, GCP, Azure...).
 
-```shell
-pip3 install -r requirements.txt
+The project provides an API offering all the primitives required to build private, context-aware AI applications.
+It follows and extends the [OpenAI API standard](https://openai.com/blog/openai-api),
+and supports both normal and streaming responses.
+
+The API is divided into two logical blocks:
+
+**High-level API**, which abstracts all the complexity of a RAG (Retrieval Augmented Generation)
+pipeline implementation:
+- Ingestion of documents: internally managing document parsing,
+splitting, metadata extraction, embedding generation and storage.
+- Chat & Completions using context from ingested documents:
+abstracting the retrieval of context, the prompt engineering and the response generation.
+
+**Low-level API**, which allows advanced users to implement their own complex pipelines:
+- Embeddings generation: based on a piece of text.
+- Contextual chunks retrieval: given a query, returns the most relevant chunks of text from the ingested documents.
+
+In addition to this, a working [Gradio UI](https://www.gradio.app/)
+client is provided to test the API, together with a set of useful tools such as bulk model
+download script, ingestion script, documents folder watch, etc.
+
+## 🎞️ Overview
+>[!WARNING]
+>  This README is not updated as frequently as the [documentation](https://docs.privategpt.dev/).
+>  Please check it out for the latest updates!
+
+### Motivation behind PrivateGPT
+Generative AI is a game changer for our society, but adoption in companies of all sizes and data-sensitive
+domains like healthcare or legal is limited by a clear concern: **privacy**.
+Not being able to ensure that your data is fully under your control when using third-party AI tools
+is a risk those industries cannot take.
+
+### Primordial version
+The first version of PrivateGPT was launched in May 2023 as a novel approach to address the privacy
+concerns by using LLMs in a complete offline way.
+
+That version, which rapidly became a go-to project for privacy-sensitive setups and served as the seed
+for thousands of local-focused generative AI projects, was the foundation of what PrivateGPT is becoming nowadays;
+thus a simpler and more educational implementation to understand the basic concepts required
+to build a fully local -and therefore, private- chatGPT-like tool.
+
+If you want to keep experimenting with it, we have saved it in the
+[primordial branch](https://github.com/zylon-ai/private-gpt/tree/primordial) of the project.
+
+> It is strongly recommended to do a clean clone and install of this new version of
+PrivateGPT if you come from the previous, primordial version.
+
+### Present and Future of PrivateGPT
+PrivateGPT is now evolving towards becoming a gateway to generative AI models and primitives, including
+completions, document ingestion, RAG pipelines and other low-level building blocks.
+We want to make it easier for any developer to build AI applications and experiences, as well as provide
+a suitable extensive architecture for the community to keep contributing.
+
+Stay tuned to our [releases](https://github.com/zylon-ai/private-gpt/releases) to check out all the new features and changes included.
+
+## 📄 Documentation
+Full documentation on installation, dependencies, configuration, running the server, deployment options,
+ingesting local documents, API details and UI features can be found here: https://docs.privategpt.dev/
+
+## 🧩 Architecture
+Conceptually, PrivateGPT is an API that wraps a RAG pipeline and exposes its
+primitives.
+* The API is built using [FastAPI](https://fastapi.tiangolo.com/) and follows
+  [OpenAI's API scheme](https://platform.openai.com/docs/api-reference).
+* The RAG pipeline is based on [LlamaIndex](https://www.llamaindex.ai/).
+
+The design of PrivateGPT allows to easily extend and adapt both the API and the
+RAG implementation. Some key architectural decisions are:
+* Dependency Injection, decoupling the different components and layers.
+* Usage of LlamaIndex abstractions such as `LLM`, `BaseEmbedding` or `VectorStore`,
+  making it immediate to change the actual implementations of those abstractions.
+* Simplicity, adding as few layers and new abstractions as possible.
+* Ready to use, providing a full implementation of the API and RAG
+  pipeline.
+
+Main building blocks:
+* APIs are defined in `private_gpt:server:<api>`. Each package contains an
+  `<api>_router.py` (FastAPI layer) and an `<api>_service.py` (the
+  service implementation). Each *Service* uses LlamaIndex base abstractions instead
+  of specific implementations,
+  decoupling the actual implementation from its usage.
+* Components are placed in
+  `private_gpt:components:<component>`. Each *Component* is in charge of providing
+  actual implementations to the base abstractions used in the Services - for example
+  `LLMComponent` is in charge of providing an actual implementation of an `LLM`
+  (for example `LlamaCPP` or `OpenAI`).
+
+## 💡 Contributing
+Contributions are welcomed! To ensure code quality we have enabled several format and
+typing checks, just run `make check` before committing to make sure your code is ok.
+Remember to test your code! You'll find a tests folder with helpers, and you can run
+tests using `make test` command.
+
+Don't know what to contribute? Here is the public 
+[Project Board](https://github.com/users/imartinez/projects/3) with several ideas. 
+
+Head over to Discord 
+#contributors channel and ask for write permissions on that GitHub project.
+
+## 💬 Community
+Join the conversation around PrivateGPT on our:
+- [Twitter (aka X)](https://twitter.com/PrivateGPT_AI)
+- [Discord](https://discord.gg/bK6mRVpErU)
+
+## 📖 Citation
+If you use PrivateGPT in a paper, check out the [Citation file](CITATION.cff) for the correct citation.  
+You can also use the "Cite this repository" button in this repo to get the citation in different formats.
+
+Here are a couple of examples:
+
+#### BibTeX
+```bibtex
+@software{Zylon_PrivateGPT_2023,
+author = {Zylon by PrivateGPT},
+license = {Apache-2.0},
+month = may,
+title = {{PrivateGPT}},
+url = {https://github.com/zylon-ai/private-gpt},
+year = {2023}
+}
 ```
 
-*Alternative requirements installation with poetry*
-1. Install [poetry](https://python-poetry.org/docs/#installation)
-
-2. Run this commands
-```shell
-cd privateGPT
-poetry install
-poetry shell
+#### APA
+```
+Zylon by PrivateGPT (2023). PrivateGPT [Computer software]. https://github.com/zylon-ai/private-gpt
 ```
 
-Then, download the LLM model and place it in a directory of your choice:
-- LLM: default to [ggml-gpt4all-j-v1.3-groovy.bin](https://gpt4all.io/models/ggml-gpt4all-j-v1.3-groovy.bin). If you prefer a different GPT4All-J compatible model, just download it and reference it in your `.env` file.
+## 🤗 Partners & Supporters
+PrivateGPT is actively supported by the teams behind:
+* [Qdrant](https://qdrant.tech/), providing the default vector database
+* [Fern](https://buildwithfern.com/), providing Documentation and SDKs
+* [LlamaIndex](https://www.llamaindex.ai/), providing the base RAG framework and abstractions
 
-Copy the `example.env` template into `.env`
-```shell
-cp example.env .env
-```
-
-and edit the variables appropriately in the `.env` file.
-```
-MODEL_TYPE: supports LlamaCpp or GPT4All
-PERSIST_DIRECTORY: is the folder you want your vectorstore in
-MODEL_PATH: Path to your GPT4All or LlamaCpp supported LLM
-MODEL_N_CTX: Maximum token limit for the LLM model
-MODEL_N_BATCH: Number of tokens in the prompt that are fed into the model at a time. Optimal value differs a lot depending on the model (8 works well for GPT4All, and 1024 is better for LlamaCpp)
-EMBEDDINGS_MODEL_NAME: SentenceTransformers embeddings model name (see https://www.sbert.net/docs/pretrained_models.html)
-TARGET_SOURCE_CHUNKS: The amount of chunks (sources) that will be used to answer a question
-```
-
-Note: because of the way `langchain` loads the `SentenceTransformers` embeddings, the first time you run the script it will require internet connection to download the embeddings model itself.
-
-## Test dataset
-This repo uses a [state of the union transcript](https://github.com/imartinez/privateGPT/blob/main/source_documents/state_of_the_union.txt) as an example.
-
-## Instructions for ingesting your own dataset
-
-Put any and all your files into the `source_documents` directory
-
-The supported extensions are:
-
-   - `.csv`: CSV,
-   - `.docx`: Word Document,
-   - `.doc`: Word Document,
-   - `.enex`: EverNote,
-   - `.eml`: Email,
-   - `.epub`: EPub,
-   - `.html`: HTML File,
-   - `.md`: Markdown,
-   - `.msg`: Outlook Message,
-   - `.odt`: Open Document Text,
-   - `.pdf`: Portable Document Format (PDF),
-   - `.pptx` : PowerPoint Document,
-   - `.ppt` : PowerPoint Document,
-   - `.txt`: Text file (UTF-8),
-
-Run the following command to ingest all the data.
-
-```shell
-python ingest.py
-```
-
-Output should look like this:
-
-```shell
-Creating new vectorstore
-Loading documents from source_documents
-Loading new documents: 100%|██████████████████████| 1/1 [00:01<00:00,  1.73s/it]
-Loaded 1 new documents from source_documents
-Split into 90 chunks of text (max. 500 tokens each)
-Creating embeddings. May take some minutes...
-Using embedded DuckDB with persistence: data will be stored in: db
-Ingestion complete! You can now run privateGPT.py to query your documents
-```
-
-It will create a `db` folder containing the local vectorstore. Will take 20-30 seconds per document, depending on the size of the document.
-You can ingest as many documents as you want, and all will be accumulated in the local embeddings database.
-If you want to start from an empty database, delete the `db` folder.
-
-Note: during the ingest process no data leaves your local environment. You could ingest without an internet connection, except for the first time you run the ingest script, when the embeddings model is downloaded.
-
-## Ask questions to your documents, locally!
-In order to ask a question, run a command like:
-
-```shell
-python privateGPT.py
-```
-
-And wait for the script to require your input.
-
-```plaintext
-> Enter a query:
-```
-
-Hit enter. You'll need to wait 20-30 seconds (depending on your machine) while the LLM model consumes the prompt and prepares the answer. Once done, it will print the answer and the 4 sources it used as context from your documents; you can then ask another question without re-running the script, just wait for the prompt again.
-
-Note: you could turn off your internet connection, and the script inference would still work. No data gets out of your local environment.
-
-Type `exit` to finish the script.
-
-
-### CLI
-The script also supports optional command-line arguments to modify its behavior. You can see a full list of these arguments by running the command ```python privateGPT.py --help``` in your terminal.
-
-
-# How does it work?
-Selecting the right local models and the power of `LangChain` you can run the entire pipeline locally, without any data leaving your environment, and with reasonable performance.
-
-- `ingest.py` uses `LangChain` tools to parse the document and create embeddings locally using `HuggingFaceEmbeddings` (`SentenceTransformers`). It then stores the result in a local vector database using `Chroma` vector store.
-- `privateGPT.py` uses a local LLM based on `GPT4All-J` or `LlamaCpp` to understand questions and create answers. The context for the answers is extracted from the local vector store using a similarity search to locate the right piece of context from the docs.
-- `GPT4All-J` wrapper was introduced in LangChain 0.0.162.
-
-# System Requirements
-
-## Python Version
-To use this software, you must have Python 3.10 or later installed. Earlier versions of Python will not compile.
-
-## C++ Compiler
-If you encounter an error while building a wheel during the `pip install` process, you may need to install a C++ compiler on your computer.
-
-### For Windows 10/11
-To install a C++ compiler on Windows 10/11, follow these steps:
-
-1. Install Visual Studio 2022.
-2. Make sure the following components are selected:
-   * Universal Windows Platform development
-   * C++ CMake tools for Windows
-3. Download the MinGW installer from the [MinGW website](https://sourceforge.net/projects/mingw/).
-4. Run the installer and select the `gcc` component.
-
-## Mac Running Intel
-When running a Mac with Intel hardware (not M1), you may run into _clang: error: the clang compiler does not support '-march=native'_ during pip install.
-
-If so set your archflags during pip install. eg: _ARCHFLAGS="-arch x86_64" pip3 install -r requirements.txt_
-
-# Disclaimer
-This is a test project to validate the feasibility of a fully private solution for question answering using LLMs and Vector embeddings. It is not production ready, and it is not meant to be used in production. The models selection is not optimized for performance, but for privacy; but it is possible to use different models and vectorstores to improve performance.
+This project has been strongly influenced and supported by other amazing projects like 
+[LangChain](https://github.com/hwchase17/langchain),
+[GPT4All](https://github.com/nomic-ai/gpt4all),
+[LlamaCpp](https://github.com/ggerganov/llama.cpp),
+[Chroma](https://www.trychroma.com/)
+and [SentenceTransformers](https://www.sbert.net/).

constants.py

@@ -1,16 +0,0 @@
-import os
-from dotenv import load_dotenv
-from chromadb.config import Settings
-
-load_dotenv()
-
-# Define the folder for storing database
-PERSIST_DIRECTORY = os.environ.get('PERSIST_DIRECTORY')
-if PERSIST_DIRECTORY is None:
-    raise Exception("Please set the PERSIST_DIRECTORY environment variable")
-
-# Define the Chroma settings
-CHROMA_SETTINGS = Settings(
-        persist_directory=PERSIST_DIRECTORY,
-        anonymized_telemetry=False
-)

docker-compose.yaml

@@ -0,0 +1,102 @@
+services:
+
+  #-----------------------------------
+  #---- Private-GPT services ---------
+  #-----------------------------------
+
+  # Private-GPT service for the Ollama CPU and GPU modes
+  # This service builds from an external Dockerfile and runs the Ollama mode.
+  private-gpt-ollama:
+    image: ${PGPT_IMAGE:-zylonai/private-gpt}${PGPT_TAG:-0.6.1}-ollama
+    build:
+      context: .
+      dockerfile: Dockerfile.ollama
+    volumes:
+      - ./local_data/:/home/worker/app/local_data
+    ports:
+      - "8001:8001"
+    environment:
+      PORT: 8001
+      PGPT_PROFILES: docker
+      PGPT_MODE: ollama
+      PGPT_EMBED_MODE: ollama
+      PGPT_OLLAMA_API_BASE: http://ollama:11434
+      HF_TOKEN: ${HF_TOKEN:-}
+    profiles:
+      - ""
+      - ollama-cpu
+      - ollama-cuda
+      - ollama-api
+
+  # Private-GPT service for the local mode
+  # This service builds from a local Dockerfile and runs the application in local mode.
+  private-gpt-llamacpp-cpu:
+    image: ${PGPT_IMAGE:-zylonai/private-gpt}${PGPT_TAG:-0.6.1}-llamacpp-cpu
+    build:
+      context: .
+      dockerfile: Dockerfile.llamacpp-cpu
+    volumes:
+      - ./local_data/:/home/worker/app/local_data
+      - ./models/:/home/worker/app/models
+    entrypoint: sh -c ".venv/bin/python scripts/setup && .venv/bin/python -m private_gpt"
+    ports:
+      - "8001:8001"
+    environment:
+      PORT: 8001
+      PGPT_PROFILES: local
+      HF_TOKEN: ${HF_TOKEN}
+    profiles:
+      - llamacpp-cpu
+
+  #-----------------------------------
+  #---- Ollama services --------------
+  #-----------------------------------
+
+  # Traefik reverse proxy for the Ollama service
+  # This will route requests to the Ollama service based on the profile.
+  ollama:
+    image: traefik:v2.10
+    ports:
+      - "11435:11434"
+      - "8081:8080"
+    command:
+      - "--providers.file.filename=/etc/router.yml"
+      - "--log.level=ERROR"
+      - "--api.insecure=true"
+      - "--providers.docker=true"
+      - "--providers.docker.exposedbydefault=false"
+      - "--entrypoints.web.address=:11434"
+    volumes:
+      - /var/run/docker.sock:/var/run/docker.sock:ro
+      - ./.docker/router.yml:/etc/router.yml:ro
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
+    profiles:
+      - ""
+      - ollama-cpu
+      - ollama-cuda
+      - ollama-api
+
+  # Ollama service for the CPU mode
+  ollama-cpu:
+    image: ollama/ollama:latest
+    volumes:
+      - ./models:/root/.ollama
+    profiles:
+      - ""
+      - ollama
+
+  # Ollama service for the CUDA mode
+  ollama-cuda:
+    image: ollama/ollama:latest
+    volumes:
+      - ./models:/root/.ollama
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]
+    profiles:
+      - ollama-cuda

example.env

@@ -1,7 +0,0 @@
-PERSIST_DIRECTORY=db
-MODEL_TYPE=GPT4All
-MODEL_PATH=models/ggml-gpt4all-j-v1.3-groovy.bin
-EMBEDDINGS_MODEL_NAME=all-MiniLM-L6-v2
-MODEL_N_CTX=1000
-MODEL_N_BATCH=8
-TARGET_SOURCE_CHUNKS=4

fern/README.md

@@ -0,0 +1,39 @@
+# Documentation of PrivateGPT
+
+The documentation of this project is being rendered thanks to [fern](https://github.com/fern-api/fern).
+
+Fern is basically transforming your `.md` and `.mdx` files into a static website: your documentation.
+
+The configuration of your documentation is done in the `./docs.yml` file.
+There, you can configure the navbar, tabs, sections and pages being rendered.
+
+The documentation of fern (and the syntax of its configuration `docs.yml`) is 
+available there [docs.buildwithfern.com](https://docs.buildwithfern.com/).
+
+## How to run fern
+
+**You cannot render your documentation locally without fern credentials.**
+
+To see how your documentation looks like, you **have to** use the CICD of this
+repository (by opening a PR, CICD job will be executed, and a preview of 
+your PR's documentation will be deployed in vercel automatically, through fern).
+
+The only thing you can do locally, is to run `fern check`, which check the syntax of
+your `docs.yml` file.
+
+## How to add a new page
+Add in the `docs.yml` a new `page`, with the following syntax:
+
+```yml
+navigation:
+  # ...
+  - tab: my-existing-tab
+    layout:
+      # ...
+      - section: My Existing Section
+        contents:
+          # ...
+          - page: My new page display name
+            # The path of the page, relative to `fern/`
+            path: ./docs/pages/my-existing-tab/new-page-content.mdx
+```

fern/docs.yml

@@ -0,0 +1,129 @@
+# Main Fern configuration file
+instances:
+  - url: privategpt.docs.buildwithfern.com
+    custom-domain: docs.privategpt.dev
+
+title: PrivateGPT | Docs
+
+# The tabs definition, in the top left corner
+tabs:
+  overview:
+    display-name: Overview
+    icon: "fa-solid fa-home"
+  quickstart:
+    display-name: Quickstart
+    icon: "fa-solid fa-rocket"
+  installation:
+    display-name: Installation
+    icon: "fa-solid fa-download"
+  manual:
+    display-name: Manual
+    icon: "fa-solid fa-book"
+  recipes:
+    display-name: Recipes
+    icon: "fa-solid fa-flask"
+  api-reference:
+    display-name: API Reference
+    icon: "fa-solid fa-file-contract"
+
+# Definition of tabs contents, will be displayed on the left side of the page, below all tabs
+navigation:
+  # The default tab
+  - tab: overview
+    layout:
+      - section: Welcome
+        contents:
+          - page: Introduction
+            path: ./docs/pages/overview/welcome.mdx
+  - tab: quickstart
+    layout:
+      - section: Getting started
+        contents:
+          - page: Quickstart
+            path: ./docs/pages/quickstart/quickstart.mdx
+  # How to install PrivateGPT, with FAQ and troubleshooting
+  - tab: installation
+    layout:
+      - section: Getting started
+        contents:
+          - page: Main Concepts
+            path: ./docs/pages/installation/concepts.mdx
+          - page: Installation
+            path: ./docs/pages/installation/installation.mdx
+          - page: Troubleshooting
+            path: ./docs/pages/installation/troubleshooting.mdx
+  # Manual of PrivateGPT: how to use it and configure it
+  - tab: manual
+    layout:
+      - section: General configuration
+        contents:
+          - page: Configuration
+            path: ./docs/pages/manual/settings.mdx
+      - section: Document management
+        contents:
+          - page: Ingestion
+            path: ./docs/pages/manual/ingestion.mdx
+          - page: Deletion
+            path: ./docs/pages/manual/ingestion-reset.mdx
+      - section: Storage
+        contents:
+          - page: Vector Stores
+            path: ./docs/pages/manual/vectordb.mdx
+          - page: Node Stores
+            path: ./docs/pages/manual/nodestore.mdx
+      - section: Advanced Setup
+        contents:
+          - page: LLM Backends
+            path: ./docs/pages/manual/llms.mdx
+          - page: Reranking
+            path: ./docs/pages/manual/reranker.mdx
+      - section: User Interface
+        contents:
+          - page: Gradio Manual
+            path: ./docs/pages/ui/gradio.mdx
+          - page: Alternatives
+            path: ./docs/pages/ui/alternatives.mdx
+  - tab: recipes
+    layout:
+      - section: Getting started
+        contents:
+          - page: Quickstart
+            path: ./docs/pages/recipes/quickstart.mdx
+      - section: General use cases
+        contents:
+          - page: Summarize
+            path: ./docs/pages/recipes/summarize.mdx
+  # More advanced usage of PrivateGPT, by API
+  - tab: api-reference
+    layout:
+      - section: Overview
+        contents:
+          - page : API Reference overview
+            path: ./docs/pages/api-reference/api-reference.mdx
+          - page: SDKs
+            path: ./docs/pages/api-reference/sdks.mdx
+      - api: API Reference
+
+# Definition of the navbar, will be displayed in the top right corner.
+# `type:primary` is always displayed at the most right side of the navbar
+navbar-links:
+  - type: secondary
+    text: Contact us
+    url: "mailto:hello@zylon.ai"
+  - type: github
+    value: "https://github.com/zylon-ai/private-gpt"
+  - type: primary
+    text: Join the Discord
+    url: https://discord.com/invite/bK6mRVpErU
+
+colors:
+  accentPrimary:
+    dark: "#C6BBFF"
+    light: "#756E98"
+
+logo:
+  dark: ./docs/assets/logo_light.png
+  light: ./docs/assets/logo_dark.png
+  height: 50
+
+favicon: ./docs/assets/favicon.ico

fern/docs/pages/api-reference/api-reference.mdx

@@ -0,0 +1,14 @@
+# API Reference
+
+The API is divided in two logical blocks:
+
+1. High-level API, abstracting all the complexity of a RAG (Retrieval Augmented Generation) pipeline implementation:
+    - Ingestion of documents: internally managing document parsing, splitting, metadata extraction,
+      embedding generation and storage.
+    - Chat & Completions using context from ingested documents: abstracting the retrieval of context, the prompt
+      engineering and the response generation.
+
+2. Low-level API, allowing advanced users to implement their own complex pipelines:
+    - Embeddings generation: based on a piece of text.
+    - Contextual chunks retrieval: given a query, returns the most relevant chunks of text from the ingested
+      documents.

fern/docs/pages/api-reference/sdks.mdx

@@ -0,0 +1,38 @@
+We use [Fern](www.buildwithfern.com) to offer API clients for Node.js, Python, Go, and Java.
+We recommend using these clients to interact with our endpoints.
+The clients are kept up to date automatically, so we encourage you to use the latest version.
+
+## SDKs
+
+*Coming soon!*
+
+<Cards>
+  <Card
+    title="TypeScript"
+    icon="fa-brands fa-node"
+    href="https://github.com/zylon-ai/privategpt-ts"
+  />
+  <Card
+    title="Python"
+    icon="fa-brands fa-python"
+    href="https://github.com/zylon-ai/pgpt-python"
+  />
+  <br />
+</Cards>
+
+<br />
+
+<Cards>
+  <Card
+    title="Java - WIP"
+    icon="fa-brands fa-java"
+    href="https://github.com/zylon-ai/private-gpt-java"
+  />
+  <Card
+    title="Go - WIP"
+    icon="fa-brands fa-golang"
+    href="https://github.com/zylon-ai/private-gpt-go"
+  />
+</Cards>
+
+<br />

fern/docs/pages/installation/concepts.mdx

@@ -0,0 +1,67 @@
+PrivateGPT is a service that wraps a set of AI RAG primitives in a comprehensive set of APIs providing a private, secure, customizable and easy to use GenAI development framework.
+
+It uses FastAPI and LLamaIndex as its core frameworks. Those can be customized by changing the codebase itself.
+
+It supports a variety of LLM providers, embeddings providers, and vector stores, both local and remote. Those can be easily changed without changing the codebase.
+
+# Different Setups support
+
+## Setup configurations available
+You get to decide the setup for these 3 main components:
+- **LLM**: the large language model provider used for inference. It can be local, or remote, or even OpenAI.
+- **Embeddings**: the embeddings provider used to encode the input, the documents and the users' queries. Same as the LLM, it can be local, or remote, or even OpenAI.
+- **Vector store**: the store used to index and retrieve the documents.
+
+There is an extra component that can be enabled or disabled: the UI. It is a Gradio UI that allows to interact with the API in a more user-friendly way.
+
+<Callout intent = "warning">
+A working **Gradio UI client** is provided to test the API, together with a set of useful tools such as bulk
+model download script, ingestion script, documents folder watch, etc. Please refer to the [UI alternatives](/manual/user-interface/alternatives) page for more UI alternatives.
+</Callout>
+
+### Setups and Dependencies
+Your setup will be the combination of the different options available. You'll find recommended setups in the [installation](./installation) section.
+PrivateGPT uses poetry to manage its dependencies. You can install the dependencies for the different setups by running `poetry install --extras "<extra1> <extra2>..."`.
+Extras are the different options available for each component. For example, to install the dependencies for a a local setup with UI and qdrant as vector database, Ollama as LLM and local embeddings, you would run:
+
+```bash
+poetry install --extras "ui vector-stores-qdrant llms-ollama embeddings-ollama"
+```
+
+Refer to the [installation](./installation) section for more details.
+
+### Setups and Configuration
+PrivateGPT uses yaml to define its configuration in files named `settings-<profile>.yaml`.
+Different configuration files can be created in the root directory of the project.
+PrivateGPT will load the configuration at startup from the profile specified in the `PGPT_PROFILES` environment variable.
+For example, running:
+```bash
+PGPT_PROFILES=ollama make run
+```
+will load the configuration from `settings.yaml` and `settings-ollama.yaml`.
+- `settings.yaml` is always loaded and contains the default configuration.
+- `settings-ollama.yaml` is loaded if the `ollama` profile is specified in the `PGPT_PROFILES` environment variable. It can override configuration from the default `settings.yaml`
+
+## About Fully Local Setups
+In order to run PrivateGPT in a fully local setup, you will need to run the LLM, Embeddings and Vector Store locally.
+
+### LLM
+For local LLM there are two options:
+* (Recommended) You can use the 'ollama' option in PrivateGPT, which will connect to your local Ollama instance. Ollama simplifies a lot the installation of local LLMs.
+* You can use the 'llms-llama-cpp' option in PrivateGPT, which will use LlamaCPP. It works great on Mac with Metal most of the times (leverages Metal GPU), but it can be tricky in certain Linux and Windows distributions, depending on the GPU. In the installation document you'll find guides and troubleshooting.
+
+In order for LlamaCPP powered LLM to work (the second option), you need to download the LLM model to the `models` folder. You can do so by running the `setup` script:
+```bash
+poetry run python scripts/setup
+```
+### Embeddings
+For local Embeddings there are two options:
+* (Recommended) You can use the 'ollama' option in PrivateGPT, which will connect to your local Ollama instance. Ollama simplifies a lot the installation of local LLMs.
+* You can use the 'embeddings-huggingface' option in PrivateGPT, which will use HuggingFace.
+
+In order for HuggingFace LLM to work (the second option), you need to download the embeddings model to the `models` folder. You can do so by running the `setup` script:
+```bash
+poetry run python scripts/setup
+```
+### Vector stores
+The vector stores supported (Qdrant, Milvus, ChromaDB and Postgres) run locally by default.

fern/docs/pages/installation/installation.mdx

@@ -0,0 +1,431 @@
+It is important that you review the [Main Concepts](../concepts) section to understand the different components of PrivateGPT and how they interact with each other.
+
+## Base requirements to run PrivateGPT
+
+### 1. Clone the PrivateGPT Repository
+Clone the repository and navigate to it:
+```bash
+git clone https://github.com/zylon-ai/private-gpt
+cd private-gpt
+```
+
+### 2. Install Python 3.11
+If you do not have Python 3.11 installed, install it using a Python version manager like `pyenv`. Earlier Python versions are not supported.
+#### macOS/Linux
+Install and set Python 3.11 using [pyenv](https://github.com/pyenv/pyenv):
+```bash
+pyenv install 3.11
+pyenv local 3.11
+```
+#### Windows
+Install and set Python 3.11 using [pyenv-win](https://github.com/pyenv-win/pyenv-win):
+```bash
+pyenv install 3.11
+pyenv local 3.11
+```
+
+### 3. Install `Poetry`
+Install [Poetry](https://python-poetry.org/docs/#installing-with-the-official-installer) for dependency management:
+Follow the instructions on the official Poetry website to install it.
+
+<Callout intent="warning">
+A bug exists in Poetry versions 1.7.0 and earlier. We strongly recommend upgrading to a tested version.
+To upgrade Poetry to latest tested version, run `poetry self update 1.8.3` after installing it.
+</Callout>
+
+### 4. Optional: Install `make`
+To run various scripts, you need to install `make`. Follow the instructions for your operating system:
+#### macOS
+(Using Homebrew):
+```bash
+brew install make
+```
+#### Windows
+(Using Chocolatey):
+```bash
+choco install make
+```
+
+## Install and Run Your Desired Setup
+
+PrivateGPT allows customization of the setup, from fully local to cloud-based, by deciding the modules to use. To install only the required dependencies, PrivateGPT offers different `extras` that can be combined during the installation process:
+
+```bash
+poetry install --extras "<extra1> <extra2>..."
+```
+Where `<extra>` can be any of the following options described below.
+
+### Available Modules
+
+You need to choose one option per category (LLM, Embeddings, Vector Stores, UI). Below are the tables listing the available options for each category.
+
+#### LLM
+
+| **Option**   | **Description**                                                        | **Extra**           |
+|--------------|------------------------------------------------------------------------|---------------------|
+| **ollama**   | Adds support for Ollama LLM, requires Ollama running locally           | llms-ollama         |
+| llama-cpp    | Adds support for local LLM using LlamaCPP                              | llms-llama-cpp      |
+| sagemaker    | Adds support for Amazon Sagemaker LLM, requires Sagemaker endpoints    | llms-sagemaker      |
+| openai       | Adds support for OpenAI LLM, requires OpenAI API key                   | llms-openai         |
+| openailike   | Adds support for 3rd party LLM providers compatible with OpenAI's API  | llms-openai-like    |
+| azopenai     | Adds support for Azure OpenAI LLM, requires Azure endpoints            | llms-azopenai       |
+| gemini       | Adds support for Gemini LLM, requires Gemini API key                   | llms-gemini         |
+
+#### Embeddings
+
+| **Option**       | **Description**                                                                | **Extra**               |
+|------------------|--------------------------------------------------------------------------------|-------------------------|
+| **ollama**       | Adds support for Ollama Embeddings, requires Ollama running locally            | embeddings-ollama       |
+| huggingface      | Adds support for local Embeddings using HuggingFace                            | embeddings-huggingface  |
+| openai           | Adds support for OpenAI Embeddings, requires OpenAI API key                    | embeddings-openai       |
+| sagemaker        | Adds support for Amazon Sagemaker Embeddings, requires Sagemaker endpoints     | embeddings-sagemaker    |
+| azopenai         | Adds support for Azure OpenAI Embeddings, requires Azure endpoints             | embeddings-azopenai     |
+| gemini           | Adds support for Gemini Embeddings, requires Gemini API key                    | embeddings-gemini       |
+
+#### Vector Stores
+
+| **Option**       | **Description**                         | **Extra**               |
+|------------------|-----------------------------------------|-------------------------|
+| **qdrant**       | Adds support for Qdrant vector store    | vector-stores-qdrant    |
+| milvus           | Adds support for Milvus vector store    | vector-stores-milvus    |
+| chroma           | Adds support for Chroma DB vector store | vector-stores-chroma    |
+| postgres         | Adds support for Postgres vector store  | vector-stores-postgres  |
+| clickhouse       | Adds support for Clickhouse vector store| vector-stores-clickhouse|
+
+#### UI
+
+| **Option**   | **Description**                          | **Extra** |
+|--------------|------------------------------------------|-----------|
+| Gradio       | Adds support for UI using Gradio         | ui        |
+
+<Callout intent = "warning">
+A working **Gradio UI client** is provided to test the API, together with a set of useful tools such as bulk
+model download script, ingestion script, documents folder watch, etc. Please refer to the [UI alternatives](/manual/user-interface/alternatives) page for more UI alternatives.
+</Callout>
+
+## Recommended Setups
+
+There are just some examples of recommended setups. You can mix and match the different options to fit your needs.
+You'll find more information in the Manual section of the documentation.
+
+> **Important for Windows**: In the examples below or how to run PrivateGPT with `make run`, `PGPT_PROFILES` env var is being set inline following Unix command line syntax (works on MacOS and Linux).
+If you are using Windows, you'll need to set the env var in a different way, for example:
+
+```powershell
+# Powershell
+$env:PGPT_PROFILES="ollama"
+make run
+```
+
+or
+
+```cmd
+# CMD
+set PGPT_PROFILES=ollama
+make run
+```
+
+Refer to the [troubleshooting](./troubleshooting) section for specific issues you might encounter.
+
+### Local, Ollama-powered setup - RECOMMENDED
+
+**The easiest way to run PrivateGPT fully locally** is to depend on Ollama for the LLM. Ollama provides local LLM and Embeddings super easy to install and use, abstracting the complexity of GPU support. It's the recommended setup for local development.
+
+Go to [ollama.ai](https://ollama.ai/) and follow the instructions to install Ollama on your machine.
+
+After the installation, make sure the Ollama desktop app is closed.
+
+Now, start Ollama service (it will start a local inference server, serving both the LLM and the Embeddings):
+```bash
+ollama serve
+```
+
+Install the models to be used, the default settings-ollama.yaml is configured to user llama3.1 8b LLM (~4GB) and nomic-embed-text Embeddings (~275MB)
+
+By default, PGPT will automatically pull models as needed. This behavior can be changed by modifying the `ollama.autopull_models` property.
+
+In any case, if you want to manually pull models, run the following commands:
+
+```bash
+ollama pull llama3.1
+ollama pull nomic-embed-text
+```
+
+Once done, on a different terminal, you can install PrivateGPT with the following command:
+```bash
+poetry install --extras "ui llms-ollama embeddings-ollama vector-stores-qdrant"
+```
+
+Once installed, you can run PrivateGPT. Make sure you have a working Ollama running locally before running the following command.
+
+```bash
+PGPT_PROFILES=ollama make run
+```
+
+PrivateGPT will use the already existing `settings-ollama.yaml` settings file, which is already configured to use Ollama LLM and Embeddings, and Qdrant. Review it and adapt it to your needs (different models, different Ollama port, etc.)
+
+The UI will be available at http://localhost:8001
+
+### Private, Sagemaker-powered setup
+
+If you need more performance, you can run a version of PrivateGPT that relies on powerful AWS Sagemaker machines to serve the LLM and Embeddings.
+
+You need to have access to sagemaker inference endpoints for the LLM and / or the embeddings, and have AWS credentials properly configured.
+
+Edit the `settings-sagemaker.yaml` file to include the correct Sagemaker endpoints.
+
+Then, install PrivateGPT with the following command:
+```bash
+poetry install --extras "ui llms-sagemaker embeddings-sagemaker vector-stores-qdrant"
+```
+
+Once installed, you can run PrivateGPT. Make sure you have a working Ollama running locally before running the following command.
+
+```bash
+PGPT_PROFILES=sagemaker make run
+```
+
+PrivateGPT will use the already existing `settings-sagemaker.yaml` settings file, which is already configured to use Sagemaker LLM and Embeddings endpoints, and Qdrant.
+
+The UI will be available at http://localhost:8001
+
+### Non-Private, OpenAI-powered test setup
+
+If you want to test PrivateGPT with OpenAI's LLM and Embeddings -taking into account your data is going to OpenAI!- you can run the following command:
+
+You need an OPENAI API key to run this setup.
+
+Edit the `settings-openai.yaml` file to include the correct API KEY. Never commit it! It's a secret! As an alternative to editing `settings-openai.yaml`, you can just set the env var OPENAI_API_KEY.
+
+Then, install PrivateGPT with the following command:
+```bash
+poetry install --extras "ui llms-openai embeddings-openai vector-stores-qdrant"
+```
+
+Once installed, you can run PrivateGPT.
+
+```bash
+PGPT_PROFILES=openai make run
+```
+
+PrivateGPT will use the already existing `settings-openai.yaml` settings file, which is already configured to use OpenAI LLM and Embeddings endpoints, and Qdrant.
+
+The UI will be available at http://localhost:8001
+
+### Non-Private, Azure OpenAI-powered test setup
+
+If you want to test PrivateGPT with Azure OpenAI's LLM and Embeddings -taking into account your data is going to Azure OpenAI!- you can run the following command:
+
+You need to have access to Azure OpenAI inference endpoints for the LLM and / or the embeddings, and have Azure OpenAI credentials properly configured.
+
+Edit the `settings-azopenai.yaml` file to include the correct Azure OpenAI endpoints.
+
+Then, install PrivateGPT with the following command:
+```bash
+poetry install --extras "ui llms-azopenai embeddings-azopenai vector-stores-qdrant"
+```
+
+Once installed, you can run PrivateGPT.
+
+```bash
+PGPT_PROFILES=azopenai make run
+```
+
+PrivateGPT will use the already existing `settings-azopenai.yaml` settings file, which is already configured to use Azure OpenAI LLM and Embeddings endpoints, and Qdrant.
+
+The UI will be available at http://localhost:8001
+
+### Local, Llama-CPP powered setup
+
+If you want to run PrivateGPT fully locally without relying on Ollama, you can run the following command:
+
+```bash
+poetry install --extras "ui llms-llama-cpp embeddings-huggingface vector-stores-qdrant"
+```
+
+In order for local LLM and embeddings to work, you need to download the models to the `models` folder. You can do so by running the `setup` script:
+```bash
+poetry run python scripts/setup
+```
+
+Once installed, you can run PrivateGPT with the following command:
+
+```bash
+PGPT_PROFILES=local make run
+```
+
+PrivateGPT will load the already existing `settings-local.yaml` file, which is already configured to use LlamaCPP LLM, HuggingFace embeddings and Qdrant.
+
+The UI will be available at http://localhost:8001
+
+#### Llama-CPP support
+
+For PrivateGPT to run fully locally without Ollama, Llama.cpp is required and in
+particular [llama-cpp-python](https://github.com/abetlen/llama-cpp-python)
+is used.
+
+You'll need to have a valid C++ compiler like gcc installed. See [Troubleshooting: C++ Compiler](#troubleshooting-c-compiler) for more details.
+
+> It's highly encouraged that you fully read llama-cpp and llama-cpp-python documentation relevant to your platform.
+> Running into installation issues is very likely, and you'll need to troubleshoot them yourself.
+
+##### Llama-CPP OSX GPU support
+
+You will need to build [llama.cpp](https://github.com/ggerganov/llama.cpp) with metal support.
+
+To do that, you need to install `llama.cpp` python's binding `llama-cpp-python` through pip, with the compilation flag
+that activate `METAL`: you have to pass `-DLLAMA_METAL=on` to the CMake command tha `pip` runs for you (see below).
+
+In other words, one should simply run:
+```bash
+CMAKE_ARGS="-DLLAMA_METAL=on" pip install --force-reinstall --no-cache-dir llama-cpp-python
+```
+
+The above command will force the re-installation of `llama-cpp-python` with `METAL` support by compiling
+`llama.cpp` locally with your `METAL` libraries (shipped by default with your macOS).
+
+More information is available in the documentation of the libraries themselves:
+* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python#installation-with-hardware-acceleration)
+* [llama-cpp-python's documentation](https://llama-cpp-python.readthedocs.io/en/latest/#installation-with-hardware-acceleration)
+* [llama.cpp](https://github.com/ggerganov/llama.cpp#build)
+
+##### Llama-CPP Windows NVIDIA GPU support
+
+Windows GPU support is done through CUDA.
+Follow the instructions on the original [llama.cpp](https://github.com/ggerganov/llama.cpp) repo to install the required
+dependencies.
+
+Some tips to get it working with an NVIDIA card and CUDA (Tested on Windows 10 with CUDA 11.5 RTX 3070):
+
+* Install latest VS2022 (and build tools) https://visualstudio.microsoft.com/vs/community/
+* Install CUDA toolkit https://developer.nvidia.com/cuda-downloads
+* Verify your installation is correct by running `nvcc --version` and `nvidia-smi`, ensure your CUDA version is up to
+  date and your GPU is detected.
+* [Optional] Install CMake to troubleshoot building issues by compiling llama.cpp directly https://cmake.org/download/
+
+If you have all required dependencies properly configured running the
+following powershell command should succeed.
+
+```powershell
+$env:CMAKE_ARGS='-DLLAMA_CUBLAS=on'; poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python
+```
+
+If your installation was correct, you should see a message similar to the following next
+time you start the server `BLAS = 1`.
+
+```console
+llama_new_context_with_model: total VRAM used: 4857.93 MB (model: 4095.05 MB, context: 762.87 MB)
+AVX = 1 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 0 | VSX = 0 |
+```
+
+Note that llama.cpp offloads matrix calculations to the GPU but the performance is
+still hit heavily due to latency between CPU and GPU communication. You might need to tweak
+batch sizes and other parameters to get the best performance for your particular system.
+
+##### Llama-CPP Linux NVIDIA GPU support and Windows-WSL
+
+Linux GPU support is done through CUDA.
+Follow the instructions on the original [llama.cpp](https://github.com/ggerganov/llama.cpp) repo to install the required
+external
+dependencies.
+
+Some tips:
+
+* Make sure you have an up-to-date C++ compiler
+* Install CUDA toolkit https://developer.nvidia.com/cuda-downloads
+* Verify your installation is correct by running `nvcc --version` and `nvidia-smi`, ensure your CUDA version is up to
+  date and your GPU is detected.
+
+After that running the following command in the repository will install llama.cpp with GPU support:
+
+```bash
+CMAKE_ARGS='-DLLAMA_CUBLAS=on' poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python
+```
+
+If your installation was correct, you should see a message similar to the following next
+time you start the server `BLAS = 1`.
+
+```
+llama_new_context_with_model: total VRAM used: 4857.93 MB (model: 4095.05 MB, context: 762.87 MB)
+AVX = 1 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 0 | VSX = 0 |
+```
+
+##### Llama-CPP Linux AMD GPU support
+
+Linux GPU support is done through ROCm.
+Some tips:
+* Install ROCm from [quick-start install guide](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/tutorial/quick-start.html)
+* [Install PyTorch for ROCm](https://rocm.docs.amd.com/projects/radeon/en/latest/docs/install/install-pytorch.html)
+```bash
+wget https://repo.radeon.com/rocm/manylinux/rocm-rel-6.0/torch-2.1.1%2Brocm6.0-cp311-cp311-linux_x86_64.whl
+poetry run pip install --force-reinstall --no-cache-dir torch-2.1.1+rocm6.0-cp311-cp311-linux_x86_64.whl
+```
+* Install bitsandbytes for ROCm
+```bash
+PYTORCH_ROCM_ARCH=gfx900,gfx906,gfx908,gfx90a,gfx1030,gfx1100,gfx1101,gfx940,gfx941,gfx942
+BITSANDBYTES_VERSION=62353b0200b8557026c176e74ac48b84b953a854
+git clone https://github.com/arlo-phoenix/bitsandbytes-rocm-5.6
+cd bitsandbytes-rocm-5.6
+git checkout ${BITSANDBYTES_VERSION}
+make hip ROCM_TARGET=${PYTORCH_ROCM_ARCH} ROCM_HOME=/opt/rocm/
+pip install . --extra-index-url https://download.pytorch.org/whl/nightly
+```
+
+After that running the following command in the repository will install llama.cpp with GPU support:
+```bash
+LLAMA_CPP_PYTHON_VERSION=0.2.56
+DAMDGPU_TARGETS=gfx900;gfx906;gfx908;gfx90a;gfx1030;gfx1100;gfx1101;gfx940;gfx941;gfx942
+CMAKE_ARGS="-DLLAMA_HIPBLAS=ON -DCMAKE_C_COMPILER=/opt/rocm/llvm/bin/clang -DCMAKE_CXX_COMPILER=/opt/rocm/llvm/bin/clang++ -DAMDGPU_TARGETS=${DAMDGPU_TARGETS}" poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python==${LLAMA_CPP_PYTHON_VERSION}
+```
+
+If your installation was correct, you should see a message similar to the following next time you start the server `BLAS = 1`.
+
+```
+AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 |
+```
+
+##### Llama-CPP Known issues and Troubleshooting
+
+Execution of LLMs locally still has a lot of sharp edges, specially when running on non Linux platforms.
+You might encounter several issues:
+
+* Performance: RAM or VRAM usage is very high, your computer might experience slowdowns or even crashes.
+* GPU Virtualization on Windows and OSX: Simply not possible with docker desktop, you have to run the server directly on
+  the host.
+* Building errors: Some of PrivateGPT dependencies need to build native code, and they might fail on some platforms.
+  Most likely you are missing some dev tools in your machine (updated C++ compiler, CUDA is not on PATH, etc.).
+  If you encounter any of these issues, please open an issue and we'll try to help.
+
+One of the first reflex to adopt is: get more information.
+If, during your installation, something does not go as planned, retry in *verbose* mode, and see what goes wrong.
+
+For example, when installing packages with `pip install`, you can add the option `-vvv` to show the details of the installation.
+
+##### Llama-CPP Troubleshooting: C++ Compiler
+
+If you encounter an error while building a wheel during the `pip install` process, you may need to install a C++
+compiler on your computer.
+
+**For Windows 10/11**
+
+To install a C++ compiler on Windows 10/11, follow these steps:
+
+1. Install Visual Studio 2022.
+2. Make sure the following components are selected:
+    * Universal Windows Platform development
+    * C++ CMake tools for Windows
+3. Download the MinGW installer from the [MinGW website](https://sourceforge.net/projects/mingw/).
+4. Run the installer and select the `gcc` component.
+
+**For OSX**
+
+1. Check if you have a C++ compiler installed, `Xcode` should have done it for you. To install Xcode, go to the App
+   Store and search for Xcode and install it. **Or** you can install the command line tools by running `xcode-select --install`.
+2. If not, you can install clang or gcc with homebrew `brew install gcc`
+
+##### Llama-CPP Troubleshooting: Mac Running Intel
+
+When running a Mac with Intel hardware (not M1), you may run into _clang: error: the clang compiler does not support '
+-march=native'_ during pip install.
+
+If so set your archflags during pip install. eg: _ARCHFLAGS="-arch x86_64" pip3 install -r requirements.txt_

fern/docs/pages/installation/troubleshooting.mdx

@@ -0,0 +1,49 @@
+# Downloading Gated and Private Models
+Many models are gated or private, requiring special access to use them. Follow these steps to gain access and set up your environment for using these models.
+## Accessing Gated Models
+1. **Request Access:**
+   Follow the instructions provided [here](https://huggingface.co/docs/hub/en/models-gated) to request access to the gated model.
+2. **Generate a Token:**
+   Once you have access, generate a token by following the instructions [here](https://huggingface.co/docs/hub/en/security-tokens).
+3. **Set the Token:**
+   Add the generated token to your `settings.yaml` file:
+   ```yaml
+   huggingface:
+     access_token: <your-token>
+   ```
+   Alternatively, set the `HF_TOKEN` environment variable:
+   ```bash
+   export HF_TOKEN=<your-token>
+   ```
+
+# Tokenizer Setup
+PrivateGPT uses the `AutoTokenizer` library to tokenize input text accurately. It connects to HuggingFace's API to download the appropriate tokenizer for the specified model.
+
+## Configuring the Tokenizer
+1. **Specify the Model:**
+   In your `settings.yaml` file, specify the model you want to use:
+   ```yaml
+   llm:
+     tokenizer: meta-llama/Meta-Llama-3.1-8B-Instruct
+   ```
+2. **Set Access Token for Gated Models:**
+   If you are using a gated model, ensure the `access_token` is set as mentioned in the previous section.
+This configuration ensures that PrivateGPT can download and use the correct tokenizer for the model you are working with.
+
+# Embedding dimensions mismatch
+If you encounter an error message like `Embedding dimensions mismatch`, it is likely due to the embedding model and
+current vector dimension mismatch. To resolve this issue, ensure that the model and the input data have the same vector dimensions.
+
+By default, PrivateGPT uses `nomic-embed-text` embeddings, which have a vector dimension of 768.
+If you are using a different embedding model, ensure that the vector dimensions match the model's output.
+
+<Callout intent = "warning">
+In versions below to 0.6.0, the default embedding model was `BAAI/bge-small-en-v1.5` in `huggingface` setup.
+If you plan to reuse the old generated embeddings, you need to update the `settings.yaml` file to use the correct embedding model:
+```yaml
+huggingface:
+  embedding_hf_model_name: BAAI/bge-small-en-v1.5
+embedding:
+  embed_dim: 384
+```
+</Callout>

fern/docs/pages/manual/ingestion-reset.mdx

@@ -0,0 +1,14 @@
+# Reset Local documents database
+
+When running in a local setup, you can remove all ingested documents by simply
+deleting all contents of `local_data` folder (except .gitignore).
+
+To simplify this process, you can use the command:
+```bash
+make wipe
+```
+
+# Advanced usage
+
+You can actually delete your documents from your storage by using the
+API endpoint `DELETE` in the Ingestion API.

fern/docs/pages/manual/ingestion.mdx

@@ -0,0 +1,137 @@
+# Ingesting & Managing Documents
+
+The ingestion of documents can be done in different ways:
+
+* Using the `/ingest` API
+* Using the Gradio UI
+* Using the Bulk Local Ingestion functionality (check next section)
+
+## Bulk Local Ingestion
+
+You will need to activate `data.local_ingestion.enabled` in your setting file to use this feature. Additionally,
+it is probably a good idea to set `data.local_ingestion.allow_ingest_from` to specify which folders are allowed to be ingested.
+
+<Callout intent = "warning">
+Be careful enabling this feature in a production environment, as it can be a security risk, as it allows users to
+ingest any local file with permissions.
+</Callout>
+
+When you are running PrivateGPT in a fully local setup, you can ingest a complete folder for convenience (containing
+pdf, text files, etc.)
+and optionally watch changes on it with the command:
+
+```bash
+make ingest /path/to/folder -- --watch
+```
+
+To log the processed and failed files to an additional file, use:
+
+```bash
+make ingest /path/to/folder -- --watch --log-file /path/to/log/file.log
+```
+
+**Note for Windows Users:** Depending on your Windows version and whether you are using PowerShell to execute
+PrivateGPT API calls, you may need to include the parameter name before passing the folder path for consumption:
+
+```bash
+make ingest arg=/path/to/folder -- --watch --log-file /path/to/log/file.log
+```
+
+After ingestion is complete, you should be able to chat with your documents
+by navigating to http://localhost:8001 and using the option `Query documents`,
+or using the completions / chat API.
+
+## Ingestion troubleshooting
+
+### Running out of memory
+
+To do not run out of memory, you should ingest your documents without the LLM loaded in your (video) memory.
+To do so, you should change your configuration to set `llm.mode: mock`.
+
+You can also use the existing `PGPT_PROFILES=mock` that will set the following configuration for you:
+
+```yaml
+llm:
+  mode: mock
+embedding:
+  mode: local
+```
+
+This configuration allows you to use hardware acceleration for creating embeddings while avoiding loading the full LLM into (video) memory.
+
+Once your documents are ingested, you can set the `llm.mode` value back to `local` (or your previous custom value).
+
+### Ingestion speed
+
+The ingestion speed depends on the number of documents you are ingesting, and the size of each document.
+To speed up the ingestion, you can change the ingestion mode in configuration.
+
+The following ingestion mode exist:
+* `simple`: historic behavior, ingest one document at a time, sequentially
+* `batch`: read, parse, and embed multiple documents using batches (batch read, and then batch parse, and then batch embed)
+* `parallel`: read, parse, and embed multiple documents in parallel. This is the fastest ingestion mode for local setup.
+* `pipeline`: Alternative to parallel.
+To change the ingestion mode, you can use the `embedding.ingest_mode` configuration value. The default value is `simple`.
+
+To configure the number of workers used for parallel or batched ingestion, you can use
+the `embedding.count_workers` configuration value. If you set this value too high, you might run out of
+memory, so be mindful when setting this value. The default value is `2`.
+For `batch` mode, you can easily set this value to your number of threads available on your CPU without
+running out of memory. For `parallel` mode, you should be more careful, and set this value to a lower value.
+
+The configuration below should be enough for users who want to stress more their hardware:
+```yaml
+embedding:
+  ingest_mode: parallel
+  count_workers: 4
+```
+
+If your hardware is powerful enough, and that you are loading heavy documents, you can increase the number of workers.
+It is recommended to do your own tests to find the optimal value for your hardware.
+
+If you have a `bash` shell, you can use this set of command to do your own benchmark:
+
+```bash
+# Wipe your local data, to put yourself in a clean state
+# This will delete all your ingested documents
+make wipe
+
+time PGPT_PROFILES=mock python ./scripts/ingest_folder.py ~/my-dir/to-ingest/
+```
+
+## Supported file formats
+
+PrivateGPT by default supports all the file formats that contains clear text (for example, `.txt` files, `.html`, etc.).
+However, these text based file formats as only considered as text files, and are not pre-processed in any other way.
+
+It also supports the following file formats:
+* `.hwp`
+* `.pdf`
+* `.docx`
+* `.pptx`
+* `.ppt`
+* `.pptm`
+* `.jpg`
+* `.png`
+* `.jpeg`
+* `.mp3`
+* `.mp4`
+* `.csv`
+* `.epub`
+* `.md`
+* `.mbox`
+* `.ipynb`
+* `.json`
+
+<Callout intent = "info">
+While `PrivateGPT` supports these file formats, it **might** require additional
+dependencies to be installed in your python's virtual environment.
+For example, if you try to ingest `.epub` files, `PrivateGPT` might fail to do it, and will instead display an
+explanatory error asking you to download the necessary dependencies to install this file format.
+</Callout>
+
+<Callout intent = "info">
+**Other file formats might work**, but they will be considered as plain text
+files (in other words, they will be ingested as `.txt` files).
+</Callout>
+

fern/docs/pages/manual/llms.mdx

@@ -0,0 +1,234 @@
+## Running the Server
+
+PrivateGPT supports running with different LLMs & setups.
+
+### Local models
+
+Both the LLM and the Embeddings model will run locally.
+
+Make sure you have followed the *Local LLM requirements* section before moving on.
+
+This command will start PrivateGPT using the `settings.yaml` (default profile) together with the `settings-local.yaml`
+configuration files. By default, it will enable both the API and the Gradio UI. Run:
+
+```bash
+PGPT_PROFILES=local make run
+```
+
+or
+
+```bash
+PGPT_PROFILES=local poetry run python -m private_gpt
+```
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API
+using Swagger UI.
+
+#### Customizing low level parameters
+
+Currently, not all the parameters of `llama.cpp` and `llama-cpp-python` are available at PrivateGPT's `settings.yaml` file.
+In case you need to customize parameters such as the number of layers loaded into the GPU, you might change
+these at the `llm_component.py` file under the `private_gpt/components/llm/llm_component.py`.
+
+##### Available LLM config options
+
+The `llm` section of the settings allows for the following configurations:
+
+- `mode`: how to run your llm
+- `max_new_tokens`: this lets you configure the number of new tokens the LLM will generate and add to the context window (by default Llama.cpp uses `256`)
+
+Example:
+
+```yaml
+llm:
+  mode: local
+  max_new_tokens: 256
+```
+
+If you are getting an out of memory error, you might also try a smaller model or stick to the proposed
+recommended models, instead of custom tuning the parameters.
+
+### Using OpenAI
+
+If you cannot run a local model (because you don't have a GPU, for example) or for testing purposes, you may
+decide to run PrivateGPT using OpenAI as the LLM and Embeddings model.
+
+In order to do so, create a profile `settings-openai.yaml` with the following contents:
+
+```yaml
+llm:
+  mode: openai
+
+openai:
+  api_base: <openai-api-base-url> # Defaults to https://api.openai.com/v1
+  api_key: <your_openai_api_key>  # You could skip this configuration and use the OPENAI_API_KEY env var instead
+  model: <openai_model_to_use> # Optional model to use. Default is "gpt-3.5-turbo"
+                               # Note: Open AI Models are listed here: https://platform.openai.com/docs/models
+```
+
+And run PrivateGPT loading that profile you just created:
+
+`PGPT_PROFILES=openai make run`
+
+or
+
+`PGPT_PROFILES=openai poetry run python -m private_gpt`
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.
+You'll notice the speed and quality of response is higher, given you are using OpenAI's servers for the heavy
+computations.
+
+### Using OpenAI compatible API
+
+Many tools, including [LocalAI](https://localai.io/) and [vLLM](https://docs.vllm.ai/en/latest/),
+support serving local models with an OpenAI compatible API. Even when overriding the `api_base`,
+using the `openai` mode doesn't allow you to use custom models. Instead, you should use the `openailike` mode:
+
+```yaml
+llm:
+  mode: openailike
+```
+
+This mode uses the same settings as the `openai` mode.
+
+As an example, you can follow the [vLLM quickstart guide](https://docs.vllm.ai/en/latest/getting_started/quickstart.html#openai-compatible-server)
+to run an OpenAI compatible server. Then, you can run PrivateGPT using the `settings-vllm.yaml` profile:
+
+`PGPT_PROFILES=vllm make run`
+
+### Using Azure OpenAI
+
+If you cannot run a local model (because you don't have a GPU, for example) or for testing purposes, you may
+decide to run PrivateGPT using Azure OpenAI as the LLM and Embeddings model.
+
+In order to do so, create a profile `settings-azopenai.yaml` with the following contents:
+
+```yaml
+llm:
+  mode: azopenai
+
+embedding:
+  mode: azopenai
+
+azopenai:
+  api_key: <your_azopenai_api_key>  # You could skip this configuration and use the AZ_OPENAI_API_KEY env var instead
+  azure_endpoint: <your_azopenai_endpoint> # You could skip this configuration and use the AZ_OPENAI_ENDPOINT env var instead
+  api_version: <api_version> # The API version to use. Default is "2023_05_15"
+  embedding_deployment_name: <your_embedding_deployment_name> # You could skip this configuration and use the AZ_OPENAI_EMBEDDING_DEPLOYMENT_NAME env var instead
+  embedding_model: <openai_embeddings_to_use> # Optional model to use. Default is "text-embedding-ada-002" 
+  llm_deployment_name: <your_model_deployment_name> # You could skip this configuration and use the AZ_OPENAI_LLM_DEPLOYMENT_NAME env var instead
+  llm_model: <openai_model_to_use> # Optional model to use. Default is "gpt-35-turbo"
+```
+
+And run PrivateGPT loading that profile you just created:
+
+`PGPT_PROFILES=azopenai make run`
+
+or
+
+`PGPT_PROFILES=azopenai poetry run python -m private_gpt`
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.
+You'll notice the speed and quality of response is higher, given you are using Azure OpenAI's servers for the heavy
+computations.
+
+### Using AWS Sagemaker
+
+For a fully private & performant setup, you can choose to have both your LLM and Embeddings model deployed using Sagemaker.
+
+Note: how to deploy models on Sagemaker is out of the scope of this documentation.
+
+In order to do so, create a profile `settings-sagemaker.yaml` with the following contents (remember to
+update the values of the llm_endpoint_name and embedding_endpoint_name to yours):
+
+```yaml
+llm:
+  mode: sagemaker
+
+sagemaker:
+  llm_endpoint_name: huggingface-pytorch-tgi-inference-2023-09-25-19-53-32-140
+  embedding_endpoint_name: huggingface-pytorch-inference-2023-11-03-07-41-36-479
+```
+
+And run PrivateGPT loading that profile you just created:
+
+`PGPT_PROFILES=sagemaker make run`
+
+or
+
+`PGPT_PROFILES=sagemaker poetry run python -m private_gpt`
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.
+
+### Using Ollama
+
+Another option for a fully private setup is using [Ollama](https://ollama.ai/).
+
+Note: how to deploy Ollama and pull models onto it is out of the scope of this documentation.
+
+In order to do so, create a profile `settings-ollama.yaml` with the following contents:
+
+```yaml
+llm:
+  mode: ollama
+
+ollama:
+  model: <ollama_model_to_use> # Required Model to use.
+                               # Note: Ollama Models are listed here: https://ollama.ai/library
+                               #       Be sure to pull the model to your Ollama server
+  api_base: <ollama-api-base-url> # Defaults to http://localhost:11434
+```
+
+And run PrivateGPT loading that profile you just created:
+
+`PGPT_PROFILES=ollama make run`
+
+or
+
+`PGPT_PROFILES=ollama poetry run python -m private_gpt`
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.
+
+### Using IPEX-LLM
+
+For a fully private setup on Intel GPUs (such as a local PC with an iGPU, or discrete GPUs like Arc, Flex, and Max), you can use [IPEX-LLM](https://github.com/intel-analytics/ipex-llm).
+
+To deploy Ollama and pull models using IPEX-LLM, please refer to [this guide](https://ipex-llm.readthedocs.io/en/latest/doc/LLM/Quickstart/ollama_quickstart.html). Then, follow the same steps outlined in the [Using Ollama](#using-ollama) section to create a `settings-ollama.yaml` profile and run the private-GPT server.
+
+### Using Gemini
+
+If you cannot run a local model (because you don't have a GPU, for example) or for testing purposes, you may
+decide to run PrivateGPT using Gemini as the LLM and Embeddings model. In addition, you will benefit from
+multimodal inputs, such as text and images, in a very large contextual window.
+
+In order to do so, create a profile `settings-gemini.yaml` with the following contents:
+
+```yaml
+llm:
+  mode: gemini
+
+embedding:
+  mode: gemini
+
+gemini:
+  api_key: <your_gemini_api_key>                # You could skip this configuration and use the GEMINI_API_KEY env var instead
+  model: <gemini_model_to_use>                  # Optional model to use. Default is models/gemini-pro"
+  embedding_model: <gemini_embeddings_to_use>   # Optional model to use. Default is "models/embedding-001"
+```
+
+And run PrivateGPT loading that profile you just created:
+
+`PGPT_PROFILES=gemini make run`
+
+or
+
+`PGPT_PROFILES=gemini poetry run python -m private_gpt`
+
+When the server is started it will print a log *Application startup complete*.
+Navigate to http://localhost:8001 to use the Gradio UI or to http://localhost:8001/docs (API section) to try the API.
+

fern/docs/pages/manual/nodestore.mdx

@@ -0,0 +1,66 @@
+## NodeStores
+PrivateGPT supports **Simple** and [Postgres](https://www.postgresql.org/) providers. Simple being the default.
+
+In order to select one or the other, set the `nodestore.database` property in the `settings.yaml` file to `simple` or `postgres`.
+
+```yaml
+nodestore:
+  database: simple
+```
+
+### Simple Document Store
+
+Setting up simple document store: Persist data with in-memory and disk storage.
+
+Enabling the simple document store is an excellent choice for small projects or proofs of concept where you need to persist data while maintaining minimal setup complexity. To get started, set the nodestore.database property in your settings.yaml file as follows:
+
+```yaml
+nodestore:
+  database: simple
+```
+The beauty of the simple document store is its flexibility and ease of implementation. It provides a solid foundation for managing and retrieving data without the need for complex setup or configuration. The combination of in-memory processing and disk persistence ensures that you can efficiently handle small to medium-sized datasets while maintaining data consistency across runs.
+
+### Postgres Document Store
+
+To enable Postgres, set the `nodestore.database` property in the `settings.yaml` file to `postgres` and install the `storage-nodestore-postgres` extra.  Note: Vector Embeddings Storage in Postgres is configured separately
+
+```bash
+poetry install --extras storage-nodestore-postgres
+```
+
+The available configuration options are:
+| Field         | Description                                               |
+|---------------|-----------------------------------------------------------|
+| **host**      | The server hosting the Postgres database. Default is `localhost` |
+| **port**      | The port on which the Postgres database is accessible. Default is `5432` |
+| **database**  | The specific database to connect to. Default is `postgres` |
+| **user**      | The username for database access. Default is `postgres` |
+| **password**  | The password for database access. (Required)            |
+| **schema_name** | The database schema to use. Default is `private_gpt`       |
+
+For example:
+```yaml
+nodestore:
+  database: postgres
+
+postgres:
+  host: localhost
+  port: 5432
+  database: postgres
+  user: postgres
+  password: <PASSWORD>
+  schema_name: private_gpt
+```
+
+Given the above configuration, Two PostgreSQL tables will be created upon successful connection: one for storing metadata related to the index and another for document data itself.
+
+```
+postgres=# \dt private_gpt.*
+                  List of relations
+   Schema    |      Name       | Type  |    Owner     
+-------------+-----------------+-------+--------------
+ private_gpt | data_docstore   | table | postgres
+ private_gpt | data_indexstore | table | postgres
+
+postgres=# 
+```

fern/docs/pages/manual/reranker.mdx

@@ -0,0 +1,36 @@
+## Enhancing Response Quality with Reranking
+
+PrivateGPT offers a reranking feature aimed at optimizing response generation by filtering out irrelevant documents, potentially leading to faster response times and enhanced relevance of answers generated by the LLM.
+
+### Enabling Reranking
+
+Document reranking can significantly improve the efficiency and quality of the responses by pre-selecting the most relevant documents before generating an answer. To leverage this feature, ensure that it is enabled in the RAG settings and consider adjusting the parameters to best fit your use case.
+
+#### Additional Requirements
+
+Before enabling reranking, you must install additional dependencies:
+
+```bash
+poetry install --extras rerank-sentence-transformers
+```
+
+This command installs dependencies for the cross-encoder reranker from sentence-transformers, which is currently the only supported method by PrivateGPT for document reranking.
+
+#### Configuration
+
+To enable and configure reranking, adjust the `rag` section within the `settings.yaml` file. Here are the key settings to consider:
+
+- `similarity_top_k`: Determines the number of documents to initially retrieve and consider for reranking. This value should be larger than `top_n`.
+- `rerank`:
+  - `enabled`: Set to `true` to activate the reranking feature.
+  - `top_n`: Specifies the number of documents to use in the final answer generation process, chosen from the top-ranked documents provided by `similarity_top_k`.
+
+Example configuration snippet:
+
+```yaml
+rag:
+  similarity_top_k: 10  # Number of documents to retrieve and consider for reranking
+  rerank:
+    enabled: true
+    top_n: 3  # Number of top-ranked documents to use for generating the answer
+```

fern/docs/pages/manual/settings.mdx

@@ -0,0 +1,85 @@
+# Settings and profiles for your private GPT
+
+The configuration of your private GPT server is done thanks to `settings` files (more precisely `settings.yaml`).
+These text files are written using the [YAML](https://en.wikipedia.org/wiki/YAML) syntax.
+
+While PrivateGPT is distributing safe and universal configuration files, you might want to quickly customize your
+PrivateGPT, and this can be done using the `settings` files.
+
+This project is defining the concept of **profiles** (or configuration profiles).
+This mechanism, using your environment variables, is giving you the ability to easily switch between
+configuration you've made.
+
+A typical use case of profile is to easily switch between LLM and embeddings.
+To be a bit more precise, you can change the language (to French, Spanish, Italian, English, etc) by simply changing
+the profile you've selected; no code changes required!
+
+PrivateGPT is configured through *profiles* that are defined using yaml files, and selected through env variables.
+The full list of properties configurable can be found in `settings.yaml`.
+
+## How to know which profiles exist
+Given that a profile `foo_bar` points to the file `settings-foo_bar.yaml` and vice-versa, you simply have to look
+at the files starting with `settings` and ending in `.yaml`.
+
+## How to use an existing profiles
+**Please note that the syntax to set the value of an environment variables depends on your OS**.
+You have to set environment variable `PGPT_PROFILES` to the name of the profile you want to use.
+
+For example, on **linux and macOS**, this gives:
+```bash
+export PGPT_PROFILES=my_profile_name_here
+```
+
+Windows Command Prompt (cmd) has a different syntax:
+```shell
+set PGPT_PROFILES=my_profile_name_here
+```
+
+Windows Powershell has a different syntax:
+```shell
+$env:PGPT_PROFILES="my_profile_name_here"
+```
+If the above is not working, you might want to try other ways to set an env variable in your window's terminal.
+
+---
+
+Once you've set this environment variable to the desired profile, you can simply launch your PrivateGPT,
+and it will run using your profile on top of the default configuration.
+
+## Reference
+Additional details on the profiles are described in this section
+
+### Environment variable `PGPT_SETTINGS_FOLDER`
+
+The location of the settings folder. Defaults to the root of the project.
+Should contain the default `settings.yaml` and any other `settings-{profile}.yaml`.
+
+### Environment variable `PGPT_PROFILES`
+
+By default, the profile definition in `settings.yaml` is loaded.
+Using this env var you can load additional profiles; format is a comma separated list of profile names.
+This will merge `settings-{profile}.yaml` on top of the base settings file.
+
+For example:
+`PGPT_PROFILES=local,cuda` will load `settings-local.yaml`
+and `settings-cuda.yaml`, their contents will be merged with
+later profiles properties overriding values of earlier ones like `settings.yaml`.
+
+During testing, the `test` profile will be active along with the default, therefore `settings-test.yaml`
+file is required.
+
+### Environment variables expansion
+
+Configuration files can contain environment variables,
+they will be expanded at runtime.
+
+Expansion must follow the pattern `${VARIABLE_NAME:default_value}`.
+
+For example, the following configuration will use the value of the `PORT`
+environment variable or `8001` if it's not set.
+Missing variables with no default will produce an error.
+
+```yaml
+server:
+  port: ${PORT:8001}
+```

fern/docs/pages/manual/vectordb.mdx

@@ -0,0 +1,187 @@
+## Vectorstores
+PrivateGPT supports [Qdrant](https://qdrant.tech/), [Milvus](https://milvus.io/), [Chroma](https://www.trychroma.com/), [PGVector](https://github.com/pgvector/pgvector) and [ClickHouse](https://github.com/ClickHouse/ClickHouse) as vectorstore providers. Qdrant being the default.
+
+In order to select one or the other, set the `vectorstore.database` property in the `settings.yaml` file to `qdrant`, `milvus`, `chroma`, `postgres` and `clickhouse`.
+
+```yaml
+vectorstore:
+  database: qdrant
+```
+
+### Qdrant configuration
+
+To enable Qdrant, set the `vectorstore.database` property in the `settings.yaml` file to `qdrant`.
+
+Qdrant settings can be configured by setting values to the `qdrant` property in the `settings.yaml` file.
+
+The available configuration options are:
+| Field        | Description |
+|--------------|-------------|
+| location     | If `:memory:` - use in-memory Qdrant instance. If `str` - use it as a `url` parameter.|
+| url          | Either host or str of 'Optional[scheme], host, Optional[port], Optional[prefix]'. Eg. `http://localhost:6333` |
+| port         | Port of the REST API interface. Default: `6333` |
+| grpc_port    | Port of the gRPC interface. Default: `6334` |
+| prefer_grpc  | If `true` - use gRPC interface whenever possible in custom methods. |
+| https        | If `true` - use HTTPS(SSL) protocol.|
+| api_key      | API key for authentication in Qdrant Cloud.|
+| prefix       | If set, add `prefix` to the REST URL path. Example: `service/v1` will result in `http://localhost:6333/service/v1/{qdrant-endpoint}` for REST API.|
+| timeout      | Timeout for REST and gRPC API requests. Default: 5.0 seconds for REST and unlimited for gRPC |
+| host         | Host name of Qdrant service. If url and host are not set, defaults to 'localhost'.|
+| path         | Persistence path for QdrantLocal. Eg. `local_data/private_gpt/qdrant`|
+| force_disable_check_same_thread         | Force disable check_same_thread for QdrantLocal sqlite connection, defaults to True.|
+
+By default Qdrant tries to connect to an instance of Qdrant server at `http://localhost:3000`.
+
+To obtain a local setup (disk-based database) without running a Qdrant server, configure the `qdrant.path` value in settings.yaml:
+
+```yaml
+qdrant:
+  path: local_data/private_gpt/qdrant
+```
+
+### Milvus configuration
+
+To enable Milvus, set the `vectorstore.database` property in the `settings.yaml` file to `milvus` and install the `milvus` extra.
+
+```bash
+poetry install --extras vector-stores-milvus
+```
+
+The available configuration options are:
+| Field        | Description |
+|--------------|-------------|
+| uri     | Default is set to "local_data/private_gpt/milvus/milvus_local.db" as a local file; you can also set up a more performant Milvus server on docker or k8s e.g.http://localhost:19530, as your uri; To use Zilliz Cloud, adjust the uri and token to Endpoint and Api key in Zilliz Cloud.|
+| token          | Pair with Milvus server on docker or k8s or zilliz cloud api key.|
+| collection_name         | The name of the collection, set to default "milvus_db".|
+| overwrite    | Overwrite the data in collection if it existed, set to default as True. |
+
+To obtain a local setup (disk-based database) without running a Milvus server, configure the uri value in settings.yaml, to store in local_data/private_gpt/milvus/milvus_local.db.
+
+### Chroma configuration
+
+To enable Chroma, set the `vectorstore.database` property in the `settings.yaml` file to `chroma` and install the `chroma` extra.
+
+```bash
+poetry install --extras chroma
+```
+
+By default `chroma` will use a disk-based database stored in local_data_path / "chroma_db" (being local_data_path defined in settings.yaml)
+
+### PGVector
+To use the PGVector store a [postgreSQL](https://www.postgresql.org/) database with the PGVector extension must be used.
+
+To enable PGVector, set the `vectorstore.database` property in the `settings.yaml` file to `postgres` and install the `vector-stores-postgres` extra.
+
+```bash
+poetry install --extras vector-stores-postgres
+```
+
+PGVector settings can be configured by setting values to the `postgres` property in the `settings.yaml` file.
+
+The available configuration options are:
+| Field         | Description                                               |
+|---------------|-----------------------------------------------------------|
+| **host**      | The server hosting the Postgres database. Default is `localhost` |
+| **port**      | The port on which the Postgres database is accessible. Default is `5432` |
+| **database**  | The specific database to connect to. Default is `postgres` |
+| **user**      | The username for database access. Default is `postgres` |
+| **password**  | The password for database access. (Required)            |
+| **schema_name** | The database schema to use. Default is `private_gpt`       |
+
+For example:
+```yaml
+vectorstore:
+  database: postgres
+
+postgres:
+  host: localhost
+  port: 5432
+  database: postgres
+  user: postgres
+  password: <PASSWORD>
+  schema_name: private_gpt
+```
+
+The following table will be created in the database
+```
+postgres=# \d private_gpt.data_embeddings
+                                      Table "private_gpt.data_embeddings"
+  Column   |       Type        | Collation | Nullable |                         Default
+-----------+-------------------+-----------+----------+---------------------------------------------------------
+ id        | bigint            |           | not null | nextval('private_gpt.data_embeddings_id_seq'::regclass)
+ text      | character varying |           | not null |
+ metadata_ | json              |           |          |
+ node_id   | character varying |           |          |
+ embedding | vector(768)       |           |          |
+Indexes:
+    "data_embeddings_pkey" PRIMARY KEY, btree (id)
+
+postgres=# 
+```
+The dimensions of the embeddings columns will be set based on the `embedding.embed_dim` value.  If the embedding model changes this table may need to be dropped and recreated to avoid a dimension mismatch.
+
+### ClickHouse
+
+To utilize ClickHouse as the vector store, a [ClickHouse](https://github.com/ClickHouse/ClickHouse) database must be employed.
+
+To enable ClickHouse, set the `vectorstore.database` property in the `settings.yaml` file to `clickhouse` and install the `vector-stores-clickhouse` extra.
+
+```bash
+poetry install --extras vector-stores-clickhouse
+```
+
+ClickHouse settings can be configured by setting values to the `clickhouse` property in the `settings.yaml` file.
+
+The available configuration options are:
+| Field                | Description                                                    |
+|----------------------|----------------------------------------------------------------|
+| **host**             | The server hosting the ClickHouse database. Default is `localhost` |
+| **port**             | The port on which the ClickHouse database is accessible. Default is `8123` |
+| **username**         | The username for database access. Default is `default` |
+| **password**         | The password for database access. (Optional) |
+| **database**         | The specific database to connect to. Default is `__default__` |
+| **secure**           | Use https/TLS for secure connection to the server. Default is `false` |
+| **interface**        | The protocol used for the connection, either 'http' or 'https'. (Optional) |
+| **settings**         | Specific ClickHouse server settings to be used with the session. (Optional) |
+| **connect_timeout**  | Timeout in seconds for establishing a connection. (Optional) |
+| **send_receive_timeout** | Read timeout in seconds for http connection. (Optional) |
+| **verify**           | Verify the server certificate in secure/https mode. (Optional) |
+| **ca_cert**          | Path to Certificate Authority root certificate (.pem format). (Optional) |
+| **client_cert**      | Path to TLS Client certificate (.pem format). (Optional) |
+| **client_cert_key**  | Path to the private key for the TLS Client certificate. (Optional) |
+| **http_proxy**       | HTTP proxy address. (Optional) |
+| **https_proxy**      | HTTPS proxy address. (Optional) |
+| **server_host_name** | Server host name to be checked against the TLS certificate. (Optional) |
+
+For example:
+```yaml
+vectorstore:
+  database: clickhouse
+
+clickhouse:
+  host: localhost
+  port: 8443
+  username: admin
+  password: <PASSWORD>
+  database: embeddings
+  secure: false
+```
+
+The following table will be created in the database:
+```
+clickhouse-client
+:) \d embeddings.llama_index
+                                   Table "llama_index"
+  № |  name     | type                                         | default_type | default_expression | comment | codec_expression | ttl_expression
+----|-----------|----------------------------------------------|--------------|--------------------|---------|------------------|---------------
+  1 | id        | String                                       |              |                    |         |                  |
+  2 | doc_id    | String                                       |              |                    |         |                  |
+  3 | text      | String                                       |              |                    |         |                  |
+  4 | vector    | Array(Float32)                               |              |                    |         |                  |
+  5 | node_info | Tuple(start Nullable(UInt64), end Nullable(UInt64)) |       |                    |         |                  |
+  6 | metadata  | String                                       |              |                    |         |                  |
+
+clickhouse-client
+```
+
+The dimensions of the embeddings columns will be set based on the `embedding.embed_dim` value. If the embedding model changes, this table may need to be dropped and recreated to avoid a dimension mismatch.

fern/docs/pages/overview/welcome.mdx

@@ -0,0 +1,42 @@
+PrivateGPT provides an **API** containing all the building blocks required to
+build **private, context-aware AI applications**.
+
+<Callout intent = "tip">
+If you are looking for an **enterprise-ready, fully private AI workspace**
+check out [Zylon's website](https://zylon.ai)  or [request a demo](https://cal.com/zylon/demo?source=pgpt-docs).
+Crafted by the team behind PrivateGPT, Zylon is a best-in-class AI collaborative
+workspace that can be easily deployed on-premise (data center, bare metal...) or in your private cloud (AWS, GCP, Azure...).
+</Callout>
+
+The API follows and extends OpenAI API standard, and supports both normal and streaming responses.
+That means that, if you can use OpenAI API in one of your tools, you can use your own PrivateGPT API instead,
+with no code changes, **and for free** if you are running PrivateGPT in a `local` setup.
+
+Get started by understanding the [Main Concepts and Installation](/installation) and then dive into the [API Reference](/api-reference).
+
+## Frequently Visited Resources
+
+<Cards>
+  <Card
+    title="Main Concepts"
+    icon="fa-solid fa-lines-leaning"
+    href="/installation"
+  />
+  <Card
+    title="API Reference"
+    icon="fa-solid fa-code"
+    href="/api-reference"
+  />
+  <Card
+    title="Twitter"
+    icon="fa-brands fa-twitter"
+    href="https://twitter.com/PrivateGPT_AI"
+  />
+  <Card
+    title="Discord Server"
+    icon="fa-brands fa-discord"
+    href="https://discord.gg/bK6mRVpErU"
+  />
+</Cards>
+
+<br />

fern/docs/pages/quickstart/quickstart.mdx

@@ -0,0 +1,105 @@
+This guide provides a quick start for running different profiles of PrivateGPT using Docker Compose.
+The profiles cater to various environments, including Ollama setups (CPU, CUDA, MacOS), and a fully local setup.
+
+By default, Docker Compose will download pre-built images from a remote registry when starting the services. However, you have the option to build the images locally if needed. Details on building Docker image locally are provided at the end of this guide.
+
+If you want to run PrivateGPT locally without Docker, refer to the [Local Installation Guide](/installation).
+
+## Prerequisites
+- **Docker and Docker Compose:** Ensure both are installed on your system.
+  [Installation Guide for Docker](https://docs.docker.com/get-docker/), [Installation Guide for Docker Compose](https://docs.docker.com/compose/install/).
+- **Clone PrivateGPT Repository:** Clone the PrivateGPT repository to your machine and navigate to the directory:
+  ```sh
+  git clone https://github.com/zylon-ai/private-gpt.git
+  cd private-gpt
+  ```
+
+## Setups
+
+### Ollama Setups (Recommended)
+
+#### 1. Default/Ollama CPU
+
+**Description:**
+This profile runs the Ollama service using CPU resources. It is the standard configuration for running Ollama-based Private-GPT services without GPU acceleration.
+
+**Run:**
+To start the services using pre-built images, run:
+```sh
+docker-compose up
+```
+or with a specific profile:
+```sh
+docker-compose --profile ollama-cpu up
+```
+
+#### 2. Ollama Nvidia CUDA
+
+**Description:**
+This profile leverages GPU acceleration with CUDA support, suitable for computationally intensive tasks that benefit from GPU resources.
+
+**Requirements:**
+Ensure that your system has compatible GPU hardware and the necessary NVIDIA drivers installed. The installation process is detailed [here](https://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html).
+
+**Run:**
+To start the services with CUDA support using pre-built images, run:
+```sh
+docker-compose --profile ollama-cuda up
+```
+
+#### 3. Ollama External API
+
+**Description:**
+This profile is designed for running PrivateGPT using Ollama installed on the host machine. This setup is particularly useful for MacOS users, as Docker does not yet support Metal GPU.
+
+**Requirements:**
+Install Ollama on your machine by following the instructions at [ollama.ai](https://ollama.ai/).
+
+**Run:**
+To start the Ollama service, use:
+```sh
+OLLAMA_HOST=0.0.0.0 ollama serve
+```
+To start the services with the host configuration using pre-built images, run:
+```sh
+docker-compose --profile ollama-api up
+```
+
+### Fully Local Setups
+
+#### 1. LlamaCPP CPU
+
+**Description:**
+This profile runs the Private-GPT services locally using `llama-cpp` and Hugging Face models.
+
+**Requirements:**
+A **Hugging Face Token (HF_TOKEN)** is required for accessing Hugging Face models. Obtain your token following [this guide](/installation/getting-started/troubleshooting#downloading-gated-and-private-models).
+
+**Run:**
+Start the services with your Hugging Face token using pre-built images:
+```sh
+HF_TOKEN=<your_hf_token> docker-compose --profile llamacpp-cpu up
+```
+Replace `<your_hf_token>` with your actual Hugging Face token.
+
+## Building Locally
+
+If you prefer to build Docker images locally, which is useful when making changes to the codebase or the Dockerfiles, follow these steps:
+
+### Building Locally
+To build the Docker images locally, navigate to the cloned repository directory and run:
+```sh
+docker-compose build
+```
+This command compiles the necessary Docker images based on the current codebase and Dockerfile configurations.
+
+### Forcing a Rebuild with --build
+If you have made changes and need to ensure these changes are reflected in the Docker images, you can force a rebuild before starting the services:
+```sh
+docker-compose up --build
+```
+or with a specific profile:
+```sh
+docker-compose --profile <profile_name> up --build
+```
+Replace `<profile_name>` with the desired profile.

fern/docs/pages/recipes/quickstart.mdx

@@ -0,0 +1,23 @@
+# Recipes
+
+Recipes are predefined use cases that help users solve very specific tasks using PrivateGPT.
+They provide a streamlined approach to achieve common goals with the platform, offering both a starting point and inspiration for further exploration.
+The main goal of Recipes is to empower the community to create and share solutions, expanding the capabilities of PrivateGPT.
+
+## How to Create a New Recipe
+
+1. **Identify the Task**: Define a specific task or problem that the Recipe will address.
+2. **Develop the Solution**: Create a clear and concise guide, including any necessary code snippets or configurations.
+3. **Submit a PR**: Fork the PrivateGPT repository, add your Recipe to the appropriate section, and submit a PR for review.
+
+We encourage you to be creative and think outside the box! Your contributions help shape the future of PrivateGPT.
+
+## Available Recipes
+
+<Cards>
+  <Card
+    title="Summarize"
+    icon="fa-solid fa-file-alt"
+    href="/recipes/general-use-cases/summarize"
+  />
+</Cards>

fern/docs/pages/recipes/summarize.mdx

@@ -0,0 +1,20 @@
+The Summarize Recipe provides a method to extract concise summaries from ingested documents or texts using PrivateGPT.
+This tool is particularly useful for quickly understanding large volumes of information by distilling key points and main ideas.
+
+## Use Case
+
+The primary use case for the `Summarize` tool is to automate the summarization of lengthy documents,
+making it easier for users to grasp the essential information without reading through entire texts.
+This can be applied in various scenarios, such as summarizing research papers, news articles, or business reports.
+
+## Key Features
+
+1. **Ingestion-compatible**: The user provides the text to be summarized. The text can be directly inputted or retrieved from ingested documents within the system.
+2. **Customization**: The summary generation can be influenced by providing specific `instructions` or a `prompt`. These inputs guide the model on how to frame the summary, allowing for customization according to user needs.
+3. **Streaming Support**: The tool supports streaming, allowing for real-time summary generation, which can be particularly useful for handling large texts or providing immediate feedback.
+
+## Contributing
+
+If you have ideas for improving the Summarize or want to add new features, feel free to contribute!
+You can submit your enhancements via a pull request on our [GitHub repository](https://github.com/zylon-ai/private-gpt).
+

fern/docs/pages/ui/alternatives.mdx

@@ -0,0 +1,21 @@
+
+This page aims to present different user interface (UI) alternatives for integrating and using PrivateGPT. These alternatives range from demo applications to fully customizable UI setups that can be adapted to your specific needs.
+
+**Do you have any working demo project using PrivateGPT?**
+
+Please open a PR to add it to the list, and come on our Discord to tell us about it!
+
+<Callout intent = "note">
+WIP: This page provides an overview of one of the UI alternatives available for PrivateGPT. More alternatives will be added to this page as they become available.
+</Callout>
+
+## [PrivateGPT SDK Demo App](https://github.com/frgarciames/privategpt-react)
+
+The PrivateGPT SDK demo app is a robust starting point for developers looking to integrate and customize PrivateGPT in their applications. Leveraging modern technologies like Tailwind, shadcn/ui, and Biomejs, it provides a smooth development experience and a highly customizable user interface. Refer to the [repository](https://github.com/frgarciames/privategpt-react) for more details and to get started.
+
+**Tech Stack:**
+
+- **Tailwind:** A utility-first CSS framework for rapid UI development.
+- **shadcn/ui:** A set of high-quality, customizable UI components.
+- **PrivateGPT Web SDK:** The core SDK for interacting with PrivateGPT.
+- **Biomejs formatter/linter:** A tool for maintaining code quality and consistency.

fern/docs/pages/ui/gradio.mdx

@@ -0,0 +1,71 @@
+## Gradio UI user manual
+
+Gradio UI is a ready to use way of testing most of PrivateGPT API functionalities.
+
+![Gradio PrivateGPT](https://github.com/zylon-ai/private-gpt/raw/main/fern/docs/assets/ui.png?raw=true)
+
+<Callout intent = "warning">
+A working **Gradio UI client** is provided to test the API, together with a set of useful tools such as bulk
+model download script, ingestion script, documents folder watch, etc. Please refer to the [UI alternatives](/manual/user-interface/alternatives) page for more UI alternatives.
+</Callout>
+
+### Execution Modes
+
+It has 3 modes of execution (you can select in the top-left):
+
+* Query Docs: uses the context from the
+  ingested documents to answer the questions posted in the chat. It also takes
+  into account previous chat messages as context.
+    * Makes use of `/chat/completions` API with `use_context=true` and no
+      `context_filter`.
+* Search in Docs: fast search that returns the 4 most related text
+  chunks, together with their source document and page.
+    * Makes use of `/chunks` API with no `context_filter`, `limit=4` and
+      `prev_next_chunks=0`.
+* LLM Chat: simple, non-contextual chat with the LLM. The ingested documents won't
+  be taken into account, only the previous messages.
+    * Makes use of `/chat/completions` API with `use_context=false`.
+
+### Document Ingestion
+
+Ingest documents by using the `Upload a File` button. You can check the progress of
+the ingestion in the console logs of the server.
+
+The list of ingested files is shown below the button.
+
+If you want to delete the ingested documents, refer to *Reset Local documents
+database* section in the documentation.
+
+### Chat
+
+Normal chat interface, self-explanatory ;)
+
+#### System Prompt
+You can view and change the system prompt being passed to the LLM by clicking "Additional Inputs"
+in the chat interface. The system prompt is also logged on the server.
+
+By default, the `Query Docs` mode uses the setting value `ui.default_query_system_prompt`.
+
+The `LLM Chat` mode attempts to use the optional settings value `ui.default_chat_system_prompt`.
+
+If no system prompt is entered, the UI will display the default system prompt being used
+for the active mode.
+
+##### System Prompt Examples:
+
+The system prompt can effectively provide your chat bot specialized roles, and results tailored to the prompt
+you have given the model. Examples of system prompts can be be found
+[here](https://www.w3schools.com/gen_ai/chatgpt-3-5/chatgpt-3-5_roles.php).
+
+Some interesting examples to try include:
+
+* You are -X-. You have all the knowledge and personality of -X-. Answer as if you were -X- using
+their manner of speaking and vocabulary.
+    * Example: You are Shakespeare. You have all the knowledge and personality of Shakespeare.
+    Answer as if you were Shakespeare using their manner of speaking and vocabulary.
+* You are an expert (at) -role-. Answer all questions using your expertise on -specific domain topic-.
+    * Example: You are an expert software engineer. Answer all questions using your expertise on Python.
+* You are a -role- bot, respond with -response criteria needed-. If no -response criteria- is needed,
+respond with -alternate response-.
+    * Example: You are a grammar checking bot, respond with any grammatical corrections needed. If no corrections
+    are needed, respond with "verified".

fern/fern.config.json

@@ -0,0 +1,4 @@
+{
+  "organization": "privategpt",
+  "version": "0.31.17"
+}

fern/generators.yml

@@ -0,0 +1,8 @@
+groups:
+  public:
+    generators:
+      - name: fernapi/fern-python-sdk
+        version: 0.6.2
+        output:
+          location: local-file-system
+          path: ../../pgpt-sdk/python

ingest.py

@@ -1,185 +0,0 @@
-#!/usr/bin/env python3
-import os
-import glob
-from typing import List
-from dotenv import load_dotenv
-from multiprocessing import Pool
-from tqdm import tqdm
-
-from langchain.document_loaders import (
-    CSVLoader,
-    EverNoteLoader,
-    PyMuPDFLoader,
-    TextLoader,
-    UnstructuredEmailLoader,
-    UnstructuredEPubLoader,
-    UnstructuredHTMLLoader,
-    UnstructuredMarkdownLoader,
-    UnstructuredODTLoader,
-    UnstructuredPowerPointLoader,
-    UnstructuredWordDocumentLoader,
-)
-
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.vectorstores import Chroma
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.docstore.document import Document
-
-if not load_dotenv():
-    print("Could not load .env file or it is empty. Please check if it exists and is readable.")
-    exit(1)
-
-from constants import CHROMA_SETTINGS
-import chromadb
-from chromadb.api.segment import API
-
-# Load environment variables
-persist_directory = os.environ.get('PERSIST_DIRECTORY')
-source_directory = os.environ.get('SOURCE_DIRECTORY', 'source_documents')
-embeddings_model_name = os.environ.get('EMBEDDINGS_MODEL_NAME')
-chunk_size = 500
-chunk_overlap = 50
-
-
-# Custom document loaders
-class MyElmLoader(UnstructuredEmailLoader):
-    """Wrapper to fallback to text/plain when default does not work"""
-
-    def load(self) -> List[Document]:
-        """Wrapper adding fallback for elm without html"""
-        try:
-            try:
-                doc = UnstructuredEmailLoader.load(self)
-            except ValueError as e:
-                if 'text/html content not found in email' in str(e):
-                    # Try plain text
-                    self.unstructured_kwargs["content_source"]="text/plain"
-                    doc = UnstructuredEmailLoader.load(self)
-                else:
-                    raise
-        except Exception as e:
-            # Add file_path to exception message
-            raise type(e)(f"{self.file_path}: {e}") from e
-
-        return doc
-
-
-# Map file extensions to document loaders and their arguments
-LOADER_MAPPING = {
-    ".csv": (CSVLoader, {}),
-    # ".docx": (Docx2txtLoader, {}),
-    ".doc": (UnstructuredWordDocumentLoader, {}),
-    ".docx": (UnstructuredWordDocumentLoader, {}),
-    ".enex": (EverNoteLoader, {}),
-    ".eml": (MyElmLoader, {}),
-    ".epub": (UnstructuredEPubLoader, {}),
-    ".html": (UnstructuredHTMLLoader, {}),
-    ".md": (UnstructuredMarkdownLoader, {}),
-    ".odt": (UnstructuredODTLoader, {}),
-    ".pdf": (PyMuPDFLoader, {}),
-    ".ppt": (UnstructuredPowerPointLoader, {}),
-    ".pptx": (UnstructuredPowerPointLoader, {}),
-    ".txt": (TextLoader, {"encoding": "utf8"}),
-    # Add more mappings for other file extensions and loaders as needed
-}
-
-
-def load_single_document(file_path: str) -> List[Document]:
-    ext = "." + file_path.rsplit(".", 1)[-1].lower()
-    if ext in LOADER_MAPPING:
-        loader_class, loader_args = LOADER_MAPPING[ext]
-        loader = loader_class(file_path, **loader_args)
-        return loader.load()
-
-    raise ValueError(f"Unsupported file extension '{ext}'")
-
-def load_documents(source_dir: str, ignored_files: List[str] = []) -> List[Document]:
-    """
-    Loads all documents from the source documents directory, ignoring specified files
-    """
-    all_files = []
-    for ext in LOADER_MAPPING:
-        all_files.extend(
-            glob.glob(os.path.join(source_dir, f"**/*{ext.lower()}"), recursive=True)
-        )
-        all_files.extend(
-            glob.glob(os.path.join(source_dir, f"**/*{ext.upper()}"), recursive=True)
-        )
-    filtered_files = [file_path for file_path in all_files if file_path not in ignored_files]
-
-    with Pool(processes=os.cpu_count()) as pool:
-        results = []
-        with tqdm(total=len(filtered_files), desc='Loading new documents', ncols=80) as pbar:
-            for i, docs in enumerate(pool.imap_unordered(load_single_document, filtered_files)):
-                results.extend(docs)
-                pbar.update()
-
-    return results
-
-def process_documents(ignored_files: List[str] = []) -> List[Document]:
-    """
-    Load documents and split in chunks
-    """
-    print(f"Loading documents from {source_directory}")
-    documents = load_documents(source_directory, ignored_files)
-    if not documents:
-        print("No new documents to load")
-        exit(0)
-    print(f"Loaded {len(documents)} new documents from {source_directory}")
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=chunk_size, chunk_overlap=chunk_overlap)
-    documents = text_splitter.split_documents(documents)
-    print(f"Split into {len(documents)} chunks of text (max. {chunk_size} tokens each)")
-    return documents
-
-def batch_chromadb_insertions(chroma_client: API, documents: List[Document]) -> List[Document]:
-    """
-    Split the total documents to be inserted into batches of documents that the local chroma client can process
-    """
-    # Get max batch size.
-    max_batch_size = chroma_client.max_batch_size
-    for i in range(0, len(documents), max_batch_size):
-        yield documents[i:i + max_batch_size]
-
-
-def does_vectorstore_exist(persist_directory: str, embeddings: HuggingFaceEmbeddings) -> bool:
-    """
-    Checks if vectorstore exists
-    """
-    db = Chroma(persist_directory=persist_directory, embedding_function=embeddings)
-    if not db.get()['documents']:
-        return False
-    return True
-
-def main():
-    # Create embeddings
-    embeddings = HuggingFaceEmbeddings(model_name=embeddings_model_name)
-    # Chroma client
-    chroma_client = chromadb.PersistentClient(settings=CHROMA_SETTINGS , path=persist_directory)
-
-    if does_vectorstore_exist(persist_directory, embeddings):
-        # Update and store locally vectorstore
-        print(f"Appending to existing vectorstore at {persist_directory}")
-        db = Chroma(persist_directory=persist_directory, embedding_function=embeddings, client_settings=CHROMA_SETTINGS, client=chroma_client)
-        collection = db.get()
-        documents = process_documents([metadata['source'] for metadata in collection['metadatas']])
-        print(f"Creating embeddings. May take some minutes...")
-        for batched_chromadb_insertion in batch_chromadb_insertions(chroma_client, documents):
-            db.add_documents(batched_chromadb_insertion)
-    else:
-        # Create and store locally vectorstore
-        print("Creating new vectorstore")
-        documents = process_documents()
-        print(f"Creating embeddings. May take some minutes...")
-        # Create the db with the first batch of documents to insert
-        batched_chromadb_insertions = batch_chromadb_insertions(chroma_client, documents)
-        first_insertion = next(batched_chromadb_insertions)
-        db = Chroma.from_documents(first_insertion, embeddings, persist_directory=persist_directory, client_settings=CHROMA_SETTINGS, client=chroma_client)
-        # Add the rest of batches of documents
-        for batched_chromadb_insertion in batched_chromadb_insertions:
-            db.add_documents(batched_chromadb_insertion)
-
-    print(f"Ingestion complete! You can now run privateGPT.py to query your documents")
-
-
-if __name__ == "__main__":
-    main()

local_data/.gitignore

@@ -0,0 +1,2 @@
+*
+!.gitignore

models/.gitignore

@@ -0,0 +1,2 @@
+*
+!.gitignore

privateGPT.py

@@ -1,87 +0,0 @@
-#!/usr/bin/env python3
-from dotenv import load_dotenv
-from langchain.chains import RetrievalQA
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
-from langchain.vectorstores import Chroma
-from langchain.llms import GPT4All, LlamaCpp
-import chromadb
-import os
-import argparse
-import time
-
-if not load_dotenv():
-    print("Could not load .env file or it is empty. Please check if it exists and is readable.")
-    exit(1)
-
-embeddings_model_name = os.environ.get("EMBEDDINGS_MODEL_NAME")
-persist_directory = os.environ.get('PERSIST_DIRECTORY')
-
-model_type = os.environ.get('MODEL_TYPE')
-model_path = os.environ.get('MODEL_PATH')
-model_n_ctx = os.environ.get('MODEL_N_CTX')
-model_n_batch = int(os.environ.get('MODEL_N_BATCH',8))
-target_source_chunks = int(os.environ.get('TARGET_SOURCE_CHUNKS',4))
-
-from constants import CHROMA_SETTINGS
-
-def main():
-    # Parse the command line arguments
-    args = parse_arguments()
-    embeddings = HuggingFaceEmbeddings(model_name=embeddings_model_name)
-    chroma_client = chromadb.PersistentClient(settings=CHROMA_SETTINGS , path=persist_directory)
-    db = Chroma(persist_directory=persist_directory, embedding_function=embeddings, client_settings=CHROMA_SETTINGS, client=chroma_client)
-    retriever = db.as_retriever(search_kwargs={"k": target_source_chunks})
-    # activate/deactivate the streaming StdOut callback for LLMs
-    callbacks = [] if args.mute_stream else [StreamingStdOutCallbackHandler()]
-    # Prepare the LLM
-    match model_type:
-        case "LlamaCpp":
-            llm = LlamaCpp(model_path=model_path, max_tokens=model_n_ctx, n_batch=model_n_batch, callbacks=callbacks, verbose=False)
-        case "GPT4All":
-            llm = GPT4All(model=model_path, max_tokens=model_n_ctx, backend='gptj', n_batch=model_n_batch, callbacks=callbacks, verbose=False)
-        case _default:
-            # raise exception if model_type is not supported
-            raise Exception(f"Model type {model_type} is not supported. Please choose one of the following: LlamaCpp, GPT4All")
-
-    qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents= not args.hide_source)
-    # Interactive questions and answers
-    while True:
-        query = input("\nEnter a query: ")
-        if query == "exit":
-            break
-        if query.strip() == "":
-            continue
-
-        # Get the answer from the chain
-        start = time.time()
-        res = qa(query)
-        answer, docs = res['result'], [] if args.hide_source else res['source_documents']
-        end = time.time()
-
-        # Print the result
-        print("\n\n> Question:")
-        print(query)
-        print(f"\n> Answer (took {round(end - start, 2)} s.):")
-        print(answer)
-
-        # Print the relevant sources used for the answer
-        for document in docs:
-            print("\n> " + document.metadata["source"] + ":")
-            print(document.page_content)
-
-def parse_arguments():
-    parser = argparse.ArgumentParser(description='privateGPT: Ask questions to your documents without an internet connection, '
-                                                 'using the power of LLMs.')
-    parser.add_argument("--hide-source", "-S", action='store_true',
-                        help='Use this flag to disable printing of source documents used for answers.')
-
-    parser.add_argument("--mute-stream", "-M",
-                        action='store_true',
-                        help='Use this flag to disable the streaming StdOut callback for LLMs.')
-
-    return parser.parse_args()
-
-
-if __name__ == "__main__":
-    main()

private_gpt/__init__.py

@@ -0,0 +1,27 @@
+"""private-gpt."""
+
+import logging
+import os
+
+# Set to 'DEBUG' to have extensive logging turned on, even for libraries
+ROOT_LOG_LEVEL = "INFO"
+
+PRETTY_LOG_FORMAT = (
+    "%(asctime)s.%(msecs)03d [%(levelname)-8s] %(name)+25s - %(message)s"
+)
+logging.basicConfig(level=ROOT_LOG_LEVEL, format=PRETTY_LOG_FORMAT, datefmt="%H:%M:%S")
+logging.captureWarnings(True)
+
+# Disable gradio analytics
+# This is done this way because gradio does not solely rely on what values are
+# passed to gr.Blocks(enable_analytics=...) but also on the environment
+# variable GRADIO_ANALYTICS_ENABLED. `gradio.strings` actually reads this env
+# directly, so to fully disable gradio analytics we need to set this env var.
+os.environ["GRADIO_ANALYTICS_ENABLED"] = "False"
+
+# Disable chromaDB telemetry
+# It is already disabled, see PR#1144
+# os.environ["ANONYMIZED_TELEMETRY"] = "False"
+
+# adding tiktoken cache path within repo to be able to run in offline environment.
+os.environ["TIKTOKEN_CACHE_DIR"] = "tiktoken_cache"

private_gpt/__main__.py

@@ -0,0 +1,11 @@
+# start a fastapi server with uvicorn
+
+import uvicorn
+
+from private_gpt.main import app
+from private_gpt.settings.settings import settings
+
+# Set log_config=None to do not use the uvicorn logging configuration, and
+# use ours instead. For reference, see below:
+# https://github.com/tiangolo/fastapi/discussions/7457#discussioncomment-5141108
+uvicorn.run(app, host="0.0.0.0", port=settings().server.port, log_config=None)

private_gpt/components/embedding/custom/sagemaker.py

@@ -0,0 +1,82 @@
+# mypy: ignore-errors
+import json
+from typing import Any
+
+import boto3
+from llama_index.core.base.embeddings.base import BaseEmbedding
+from pydantic import Field, PrivateAttr
+
+
+class SagemakerEmbedding(BaseEmbedding):
+    """Sagemaker Embedding Endpoint.
+
+    To use, you must supply the endpoint name from your deployed
+    Sagemaker embedding model & the region where it is deployed.
+
+    To authenticate, the AWS client uses the following methods to
+    automatically load credentials:
+    https://boto3.amazonaws.com/v1/documentation/api/latest/guide/credentials.html
+
+    If a specific credential profile should be used, you must pass
+    the name of the profile from the ~/.aws/credentials file that is to be used.
+
+    Make sure the credentials / roles used have the required policies to
+    access the Sagemaker endpoint.
+    See: https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies.html
+    """
+
+    endpoint_name: str = Field(description="")
+
+    _boto_client: Any = boto3.client(
+        "sagemaker-runtime",
+    )  # TODO make it an optional field
+
+    _async_not_implemented_warned: bool = PrivateAttr(default=False)
+
+    @classmethod
+    def class_name(cls) -> str:
+        return "SagemakerEmbedding"
+
+    def _async_not_implemented_warn_once(self) -> None:
+        if not self._async_not_implemented_warned:
+            print("Async embedding not available, falling back to sync method.")
+            self._async_not_implemented_warned = True
+
+    def _embed(self, sentences: list[str]) -> list[list[float]]:
+        request_params = {
+            "inputs": sentences,
+        }
+
+        resp = self._boto_client.invoke_endpoint(
+            EndpointName=self.endpoint_name,
+            Body=json.dumps(request_params),
+            ContentType="application/json",
+        )
+
+        response_body = resp["Body"]
+        response_str = response_body.read().decode("utf-8")
+        response_json = json.loads(response_str)
+
+        return response_json["vectors"]
+
+    def _get_query_embedding(self, query: str) -> list[float]:
+        """Get query embedding."""
+        return self._embed([query])[0]
+
+    async def _aget_query_embedding(self, query: str) -> list[float]:
+        # Warn the user that sync is being used
+        self._async_not_implemented_warn_once()
+        return self._get_query_embedding(query)
+
+    async def _aget_text_embedding(self, text: str) -> list[float]:
+        # Warn the user that sync is being used
+        self._async_not_implemented_warn_once()
+        return self._get_text_embedding(text)
+
+    def _get_text_embedding(self, text: str) -> list[float]:
+        """Get text embedding."""
+        return self._embed([text])[0]
+
+    def _get_text_embeddings(self, texts: list[str]) -> list[list[float]]:
+        """Get text embeddings."""
+        return self._embed(texts)

private_gpt/components/embedding/embedding_component.py

@@ -0,0 +1,150 @@
+import logging
+
+from injector import inject, singleton
+from llama_index.core.embeddings import BaseEmbedding, MockEmbedding
+
+from private_gpt.paths import models_cache_path
+from private_gpt.settings.settings import Settings
+
+logger = logging.getLogger(__name__)
+
+
+@singleton
+class EmbeddingComponent:
+    embedding_model: BaseEmbedding
+
+    @inject
+    def __init__(self, settings: Settings) -> None:
+        embedding_mode = settings.embedding.mode
+        logger.info("Initializing the embedding model in mode=%s", embedding_mode)
+        match embedding_mode:
+            case "huggingface":
+                try:
+                    from llama_index.embeddings.huggingface import (  # type: ignore
+                        HuggingFaceEmbedding,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Local dependencies not found, install with `poetry install --extras embeddings-huggingface`"
+                    ) from e
+
+                self.embedding_model = HuggingFaceEmbedding(
+                    model_name=settings.huggingface.embedding_hf_model_name,
+                    cache_folder=str(models_cache_path),
+                    trust_remote_code=settings.huggingface.trust_remote_code,
+                )
+            case "sagemaker":
+                try:
+                    from private_gpt.components.embedding.custom.sagemaker import (
+                        SagemakerEmbedding,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Sagemaker dependencies not found, install with `poetry install --extras embeddings-sagemaker`"
+                    ) from e
+
+                self.embedding_model = SagemakerEmbedding(
+                    endpoint_name=settings.sagemaker.embedding_endpoint_name,
+                )
+            case "openai":
+                try:
+                    from llama_index.embeddings.openai import (  # type: ignore
+                        OpenAIEmbedding,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "OpenAI dependencies not found, install with `poetry install --extras embeddings-openai`"
+                    ) from e
+
+                api_base = (
+                    settings.openai.embedding_api_base or settings.openai.api_base
+                )
+                api_key = settings.openai.embedding_api_key or settings.openai.api_key
+                model = settings.openai.embedding_model
+
+                self.embedding_model = OpenAIEmbedding(
+                    api_base=api_base,
+                    api_key=api_key,
+                    model=model,
+                )
+            case "ollama":
+                try:
+                    from llama_index.embeddings.ollama import (  # type: ignore
+                        OllamaEmbedding,
+                    )
+                    from ollama import Client  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "Local dependencies not found, install with `poetry install --extras embeddings-ollama`"
+                    ) from e
+
+                ollama_settings = settings.ollama
+
+                # Calculate embedding model. If not provided tag, it will be use latest
+                model_name = (
+                    ollama_settings.embedding_model + ":latest"
+                    if ":" not in ollama_settings.embedding_model
+                    else ollama_settings.embedding_model
+                )
+
+                self.embedding_model = OllamaEmbedding(
+                    model_name=model_name,
+                    base_url=ollama_settings.embedding_api_base,
+                )
+
+                if ollama_settings.autopull_models:
+                    if ollama_settings.autopull_models:
+                        from private_gpt.utils.ollama import (
+                            check_connection,
+                            pull_model,
+                        )
+
+                        # TODO: Reuse llama-index client when llama-index is updated
+                        client = Client(
+                            host=ollama_settings.embedding_api_base,
+                            timeout=ollama_settings.request_timeout,
+                        )
+
+                        if not check_connection(client):
+                            raise ValueError(
+                                f"Failed to connect to Ollama, "
+                                f"check if Ollama server is running on {ollama_settings.api_base}"
+                            )
+                        pull_model(client, model_name)
+
+            case "azopenai":
+                try:
+                    from llama_index.embeddings.azure_openai import (  # type: ignore
+                        AzureOpenAIEmbedding,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Azure OpenAI dependencies not found, install with `poetry install --extras embeddings-azopenai`"
+                    ) from e
+
+                azopenai_settings = settings.azopenai
+                self.embedding_model = AzureOpenAIEmbedding(
+                    model=azopenai_settings.embedding_model,
+                    deployment_name=azopenai_settings.embedding_deployment_name,
+                    api_key=azopenai_settings.api_key,
+                    azure_endpoint=azopenai_settings.azure_endpoint,
+                    api_version=azopenai_settings.api_version,
+                )
+            case "gemini":
+                try:
+                    from llama_index.embeddings.gemini import (  # type: ignore
+                        GeminiEmbedding,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Gemini dependencies not found, install with `poetry install --extras embeddings-gemini`"
+                    ) from e
+
+                self.embedding_model = GeminiEmbedding(
+                    api_key=settings.gemini.api_key,
+                    model_name=settings.gemini.embedding_model,
+                )
+            case "mock":
+                # Not a random number, is the dimensionality used by
+                # the default embedding model
+                self.embedding_model = MockEmbedding(384)

private_gpt/components/ingest/ingest_component.py

@@ -0,0 +1,517 @@
+import abc
+import itertools
+import logging
+import multiprocessing
+import multiprocessing.pool
+import os
+import threading
+from pathlib import Path
+from queue import Queue
+from typing import Any
+
+from llama_index.core.data_structs import IndexDict
+from llama_index.core.embeddings.utils import EmbedType
+from llama_index.core.indices import VectorStoreIndex, load_index_from_storage
+from llama_index.core.indices.base import BaseIndex
+from llama_index.core.ingestion import run_transformations
+from llama_index.core.schema import BaseNode, Document, TransformComponent
+from llama_index.core.storage import StorageContext
+
+from private_gpt.components.ingest.ingest_helper import IngestionHelper
+from private_gpt.paths import local_data_path
+from private_gpt.settings.settings import Settings
+from private_gpt.utils.eta import eta
+
+logger = logging.getLogger(__name__)
+
+
+class BaseIngestComponent(abc.ABC):
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        logger.debug("Initializing base ingest component type=%s", type(self).__name__)
+        self.storage_context = storage_context
+        self.embed_model = embed_model
+        self.transformations = transformations
+
+    @abc.abstractmethod
+    def ingest(self, file_name: str, file_data: Path) -> list[Document]:
+        pass
+
+    @abc.abstractmethod
+    def bulk_ingest(self, files: list[tuple[str, Path]]) -> list[Document]:
+        pass
+
+    @abc.abstractmethod
+    def delete(self, doc_id: str) -> None:
+        pass
+
+
+class BaseIngestComponentWithIndex(BaseIngestComponent, abc.ABC):
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(storage_context, embed_model, transformations, *args, **kwargs)
+
+        self.show_progress = True
+        self._index_thread_lock = (
+            threading.Lock()
+        )  # Thread lock! Not Multiprocessing lock
+        self._index = self._initialize_index()
+
+    def _initialize_index(self) -> BaseIndex[IndexDict]:
+        """Initialize the index from the storage context."""
+        try:
+            # Load the index with store_nodes_override=True to be able to delete them
+            index = load_index_from_storage(
+                storage_context=self.storage_context,
+                store_nodes_override=True,  # Force store nodes in index and document stores
+                show_progress=self.show_progress,
+                embed_model=self.embed_model,
+                transformations=self.transformations,
+            )
+        except ValueError:
+            # There are no index in the storage context, creating a new one
+            logger.info("Creating a new vector store index")
+            index = VectorStoreIndex.from_documents(
+                [],
+                storage_context=self.storage_context,
+                store_nodes_override=True,  # Force store nodes in index and document stores
+                show_progress=self.show_progress,
+                embed_model=self.embed_model,
+                transformations=self.transformations,
+            )
+            index.storage_context.persist(persist_dir=local_data_path)
+        return index
+
+    def _save_index(self) -> None:
+        self._index.storage_context.persist(persist_dir=local_data_path)
+
+    def delete(self, doc_id: str) -> None:
+        with self._index_thread_lock:
+            # Delete the document from the index
+            self._index.delete_ref_doc(doc_id, delete_from_docstore=True)
+
+            # Save the index
+            self._save_index()
+
+
+class SimpleIngestComponent(BaseIngestComponentWithIndex):
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(storage_context, embed_model, transformations, *args, **kwargs)
+
+    def ingest(self, file_name: str, file_data: Path) -> list[Document]:
+        logger.info("Ingesting file_name=%s", file_name)
+        documents = IngestionHelper.transform_file_into_documents(file_name, file_data)
+        logger.info(
+            "Transformed file=%s into count=%s documents", file_name, len(documents)
+        )
+        logger.debug("Saving the documents in the index and doc store")
+        return self._save_docs(documents)
+
+    def bulk_ingest(self, files: list[tuple[str, Path]]) -> list[Document]:
+        saved_documents = []
+        for file_name, file_data in files:
+            documents = IngestionHelper.transform_file_into_documents(
+                file_name, file_data
+            )
+            saved_documents.extend(self._save_docs(documents))
+        return saved_documents
+
+    def _save_docs(self, documents: list[Document]) -> list[Document]:
+        logger.debug("Transforming count=%s documents into nodes", len(documents))
+        with self._index_thread_lock:
+            for document in documents:
+                self._index.insert(document, show_progress=True)
+            logger.debug("Persisting the index and nodes")
+            # persist the index and nodes
+            self._save_index()
+            logger.debug("Persisted the index and nodes")
+        return documents
+
+
+class BatchIngestComponent(BaseIngestComponentWithIndex):
+    """Parallelize the file reading and parsing on multiple CPU core.
+
+    This also makes the embeddings to be computed in batches (on GPU or CPU).
+    """
+
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        count_workers: int,
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(storage_context, embed_model, transformations, *args, **kwargs)
+        # Make an efficient use of the CPU and GPU, the embedding
+        # must be in the transformations
+        assert (
+            len(self.transformations) >= 2
+        ), "Embeddings must be in the transformations"
+        assert count_workers > 0, "count_workers must be > 0"
+        self.count_workers = count_workers
+
+        self._file_to_documents_work_pool = multiprocessing.Pool(
+            processes=self.count_workers
+        )
+
+    def ingest(self, file_name: str, file_data: Path) -> list[Document]:
+        logger.info("Ingesting file_name=%s", file_name)
+        documents = IngestionHelper.transform_file_into_documents(file_name, file_data)
+        logger.info(
+            "Transformed file=%s into count=%s documents", file_name, len(documents)
+        )
+        logger.debug("Saving the documents in the index and doc store")
+        return self._save_docs(documents)
+
+    def bulk_ingest(self, files: list[tuple[str, Path]]) -> list[Document]:
+        documents = list(
+            itertools.chain.from_iterable(
+                self._file_to_documents_work_pool.starmap(
+                    IngestionHelper.transform_file_into_documents, files
+                )
+            )
+        )
+        logger.info(
+            "Transformed count=%s files into count=%s documents",
+            len(files),
+            len(documents),
+        )
+        return self._save_docs(documents)
+
+    def _save_docs(self, documents: list[Document]) -> list[Document]:
+        logger.debug("Transforming count=%s documents into nodes", len(documents))
+        nodes = run_transformations(
+            documents,  # type: ignore[arg-type]
+            self.transformations,
+            show_progress=self.show_progress,
+        )
+        # Locking the index to avoid concurrent writes
+        with self._index_thread_lock:
+            logger.info("Inserting count=%s nodes in the index", len(nodes))
+            self._index.insert_nodes(nodes, show_progress=True)
+            for document in documents:
+                self._index.docstore.set_document_hash(
+                    document.get_doc_id(), document.hash
+                )
+            logger.debug("Persisting the index and nodes")
+            # persist the index and nodes
+            self._save_index()
+            logger.debug("Persisted the index and nodes")
+        return documents
+
+
+class ParallelizedIngestComponent(BaseIngestComponentWithIndex):
+    """Parallelize the file ingestion (file reading, embeddings, and index insertion).
+
+    This use the CPU and GPU in parallel (both running at the same time), and
+    reduce the memory pressure by not loading all the files in memory at the same time.
+    """
+
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        count_workers: int,
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(storage_context, embed_model, transformations, *args, **kwargs)
+        # To make an efficient use of the CPU and GPU, the embeddings
+        # must be in the transformations (to be computed in batches)
+        assert (
+            len(self.transformations) >= 2
+        ), "Embeddings must be in the transformations"
+        assert count_workers > 0, "count_workers must be > 0"
+        self.count_workers = count_workers
+        # We are doing our own multiprocessing
+        # To do not collide with the multiprocessing of huggingface, we disable it
+        os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+        self._ingest_work_pool = multiprocessing.pool.ThreadPool(
+            processes=self.count_workers
+        )
+
+        self._file_to_documents_work_pool = multiprocessing.Pool(
+            processes=self.count_workers
+        )
+
+    def ingest(self, file_name: str, file_data: Path) -> list[Document]:
+        logger.info("Ingesting file_name=%s", file_name)
+        # Running in a single (1) process to release the current
+        # thread, and take a dedicated CPU core for computation
+        documents = self._file_to_documents_work_pool.apply(
+            IngestionHelper.transform_file_into_documents, (file_name, file_data)
+        )
+        logger.info(
+            "Transformed file=%s into count=%s documents", file_name, len(documents)
+        )
+        logger.debug("Saving the documents in the index and doc store")
+        return self._save_docs(documents)
+
+    def bulk_ingest(self, files: list[tuple[str, Path]]) -> list[Document]:
+        # Lightweight threads, used for parallelize the
+        # underlying IO calls made in the ingestion
+
+        documents = list(
+            itertools.chain.from_iterable(
+                self._ingest_work_pool.starmap(self.ingest, files)
+            )
+        )
+        return documents
+
+    def _save_docs(self, documents: list[Document]) -> list[Document]:
+        logger.debug("Transforming count=%s documents into nodes", len(documents))
+        nodes = run_transformations(
+            documents,  # type: ignore[arg-type]
+            self.transformations,
+            show_progress=self.show_progress,
+        )
+        # Locking the index to avoid concurrent writes
+        with self._index_thread_lock:
+            logger.info("Inserting count=%s nodes in the index", len(nodes))
+            self._index.insert_nodes(nodes, show_progress=True)
+            for document in documents:
+                self._index.docstore.set_document_hash(
+                    document.get_doc_id(), document.hash
+                )
+            logger.debug("Persisting the index and nodes")
+            # persist the index and nodes
+            self._save_index()
+            logger.debug("Persisted the index and nodes")
+        return documents
+
+    def __del__(self) -> None:
+        # We need to do the appropriate cleanup of the multiprocessing pools
+        # when the object is deleted. Using root logger to avoid
+        # the logger to be deleted before the pool
+        logging.debug("Closing the ingest work pool")
+        self._ingest_work_pool.close()
+        self._ingest_work_pool.join()
+        self._ingest_work_pool.terminate()
+        logging.debug("Closing the file to documents work pool")
+        self._file_to_documents_work_pool.close()
+        self._file_to_documents_work_pool.join()
+        self._file_to_documents_work_pool.terminate()
+
+
+class PipelineIngestComponent(BaseIngestComponentWithIndex):
+    """Pipeline ingestion - keeping the embedding worker pool as busy as possible.
+
+    This class implements a threaded ingestion pipeline, which comprises two threads
+    and two queues. The primary thread is responsible for reading and parsing files
+    into documents. These documents are then placed into a queue, which is
+    distributed to a pool of worker processes for embedding computation. After
+    embedding, the documents are transferred to another queue where they are
+    accumulated until a threshold is reached. Upon reaching this threshold, the
+    accumulated documents are flushed to the document store, index, and vector
+    store.
+
+    Exception handling ensures robustness against erroneous files. However, in the
+    pipelined design, one error can lead to the discarding of multiple files. Any
+    discarded files will be reported.
+    """
+
+    NODE_FLUSH_COUNT = 5000  # Save the index every # nodes.
+
+    def __init__(
+        self,
+        storage_context: StorageContext,
+        embed_model: EmbedType,
+        transformations: list[TransformComponent],
+        count_workers: int,
+        *args: Any,
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(storage_context, embed_model, transformations, *args, **kwargs)
+        self.count_workers = count_workers
+        assert (
+            len(self.transformations) >= 2
+        ), "Embeddings must be in the transformations"
+        assert count_workers > 0, "count_workers must be > 0"
+        self.count_workers = count_workers
+        # We are doing our own multiprocessing
+        # To do not collide with the multiprocessing of huggingface, we disable it
+        os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+        # doc_q stores parsed files as Document chunks.
+        # Using a shallow queue causes the filesystem parser to block
+        # when it reaches capacity. This ensures it doesn't outpace the
+        # computationally intensive embeddings phase, avoiding unnecessary
+        # memory consumption.  The semaphore is used to bound the async worker
+        # embedding computations to cause the doc Q to fill and block.
+        self.doc_semaphore = multiprocessing.Semaphore(
+            self.count_workers
+        )  # limit the doc queue to # items.
+        self.doc_q: Queue[tuple[str, str | None, list[Document] | None]] = Queue(20)
+        # node_q stores documents parsed into nodes (embeddings).
+        # Larger queue size so we don't block the embedding workers during a slow
+        # index update.
+        self.node_q: Queue[
+            tuple[str, str | None, list[Document] | None, list[BaseNode] | None]
+        ] = Queue(40)
+        threading.Thread(target=self._doc_to_node, daemon=True).start()
+        threading.Thread(target=self._write_nodes, daemon=True).start()
+
+    def _doc_to_node(self) -> None:
+        # Parse documents into nodes
+        with multiprocessing.pool.ThreadPool(processes=self.count_workers) as pool:
+            while True:
+                try:
+                    cmd, file_name, documents = self.doc_q.get(
+                        block=True
+                    )  # Documents for a file
+                    if cmd == "process":
+                        # Push CPU/GPU embedding work to the worker pool
+                        # Acquire semaphore to control access to worker pool
+                        self.doc_semaphore.acquire()
+                        pool.apply_async(
+                            self._doc_to_node_worker, (file_name, documents)
+                        )
+                    elif cmd == "quit":
+                        break
+                finally:
+                    if cmd != "process":
+                        self.doc_q.task_done()  # unblock Q joins
+
+    def _doc_to_node_worker(self, file_name: str, documents: list[Document]) -> None:
+        # CPU/GPU intensive work in its own process
+        try:
+            nodes = run_transformations(
+                documents,  # type: ignore[arg-type]
+                self.transformations,
+                show_progress=self.show_progress,
+            )
+            self.node_q.put(("process", file_name, documents, nodes))
+        finally:
+            self.doc_semaphore.release()
+            self.doc_q.task_done()  # unblock Q joins
+
+    def _save_docs(
+        self, files: list[str], documents: list[Document], nodes: list[BaseNode]
+    ) -> None:
+        try:
+            logger.info(
+                f"Saving {len(files)} files ({len(documents)} documents / {len(nodes)} nodes)"
+            )
+            self._index.insert_nodes(nodes)
+            for document in documents:
+                self._index.docstore.set_document_hash(
+                    document.get_doc_id(), document.hash
+                )
+            self._save_index()
+        except Exception:
+            # Tell the user so they can investigate these files
+            logger.exception(f"Processing files {files}")
+        finally:
+            # Clearing work, even on exception, maintains a clean state.
+            nodes.clear()
+            documents.clear()
+            files.clear()
+
+    def _write_nodes(self) -> None:
+        # Save nodes to index.  I/O intensive.
+        node_stack: list[BaseNode] = []
+        doc_stack: list[Document] = []
+        file_stack: list[str] = []
+        while True:
+            try:
+                cmd, file_name, documents, nodes = self.node_q.get(block=True)
+                if cmd in ("flush", "quit"):
+                    if file_stack:
+                        self._save_docs(file_stack, doc_stack, node_stack)
+                    if cmd == "quit":
+                        break
+                elif cmd == "process":
+                    node_stack.extend(nodes)  # type: ignore[arg-type]
+                    doc_stack.extend(documents)  # type: ignore[arg-type]
+                    file_stack.append(file_name)  # type: ignore[arg-type]
+                    # Constant saving is heavy on I/O - accumulate to a threshold
+                    if len(node_stack) >= self.NODE_FLUSH_COUNT:
+                        self._save_docs(file_stack, doc_stack, node_stack)
+            finally:
+                self.node_q.task_done()
+
+    def _flush(self) -> None:
+        self.doc_q.put(("flush", None, None))
+        self.doc_q.join()
+        self.node_q.put(("flush", None, None, None))
+        self.node_q.join()
+
+    def ingest(self, file_name: str, file_data: Path) -> list[Document]:
+        documents = IngestionHelper.transform_file_into_documents(file_name, file_data)
+        self.doc_q.put(("process", file_name, documents))
+        self._flush()
+        return documents
+
+    def bulk_ingest(self, files: list[tuple[str, Path]]) -> list[Document]:
+        docs = []
+        for file_name, file_data in eta(files):
+            try:
+                documents = IngestionHelper.transform_file_into_documents(
+                    file_name, file_data
+                )
+                self.doc_q.put(("process", file_name, documents))
+                docs.extend(documents)
+            except Exception:
+                logger.exception(f"Skipping {file_data.name}")
+        self._flush()
+        return docs
+
+
+def get_ingestion_component(
+    storage_context: StorageContext,
+    embed_model: EmbedType,
+    transformations: list[TransformComponent],
+    settings: Settings,
+) -> BaseIngestComponent:
+    """Get the ingestion component for the given configuration."""
+    ingest_mode = settings.embedding.ingest_mode
+    if ingest_mode == "batch":
+        return BatchIngestComponent(
+            storage_context=storage_context,
+            embed_model=embed_model,
+            transformations=transformations,
+            count_workers=settings.embedding.count_workers,
+        )
+    elif ingest_mode == "parallel":
+        return ParallelizedIngestComponent(
+            storage_context=storage_context,
+            embed_model=embed_model,
+            transformations=transformations,
+            count_workers=settings.embedding.count_workers,
+        )
+    elif ingest_mode == "pipeline":
+        return PipelineIngestComponent(
+            storage_context=storage_context,
+            embed_model=embed_model,
+            transformations=transformations,
+            count_workers=settings.embedding.count_workers,
+        )
+    else:
+        return SimpleIngestComponent(
+            storage_context=storage_context,
+            embed_model=embed_model,
+            transformations=transformations,
+        )

private_gpt/components/ingest/ingest_helper.py

@@ -0,0 +1,105 @@
+import logging
+from pathlib import Path
+
+from llama_index.core.readers import StringIterableReader
+from llama_index.core.readers.base import BaseReader
+from llama_index.core.readers.json import JSONReader
+from llama_index.core.schema import Document
+
+logger = logging.getLogger(__name__)
+
+
+# Inspired by the `llama_index.core.readers.file.base` module
+def _try_loading_included_file_formats() -> dict[str, type[BaseReader]]:
+    try:
+        from llama_index.readers.file.docs import (  # type: ignore
+            DocxReader,
+            HWPReader,
+            PDFReader,
+        )
+        from llama_index.readers.file.epub import EpubReader  # type: ignore
+        from llama_index.readers.file.image import ImageReader  # type: ignore
+        from llama_index.readers.file.ipynb import IPYNBReader  # type: ignore
+        from llama_index.readers.file.markdown import MarkdownReader  # type: ignore
+        from llama_index.readers.file.mbox import MboxReader  # type: ignore
+        from llama_index.readers.file.slides import PptxReader  # type: ignore
+        from llama_index.readers.file.tabular import PandasCSVReader  # type: ignore
+        from llama_index.readers.file.video_audio import (  # type: ignore
+            VideoAudioReader,
+        )
+    except ImportError as e:
+        raise ImportError("`llama-index-readers-file` package not found") from e
+
+    default_file_reader_cls: dict[str, type[BaseReader]] = {
+        ".hwp": HWPReader,
+        ".pdf": PDFReader,
+        ".docx": DocxReader,
+        ".pptx": PptxReader,
+        ".ppt": PptxReader,
+        ".pptm": PptxReader,
+        ".jpg": ImageReader,
+        ".png": ImageReader,
+        ".jpeg": ImageReader,
+        ".mp3": VideoAudioReader,
+        ".mp4": VideoAudioReader,
+        ".csv": PandasCSVReader,
+        ".epub": EpubReader,
+        ".md": MarkdownReader,
+        ".mbox": MboxReader,
+        ".ipynb": IPYNBReader,
+    }
+    return default_file_reader_cls
+
+
+# Patching the default file reader to support other file types
+FILE_READER_CLS = _try_loading_included_file_formats()
+FILE_READER_CLS.update(
+    {
+        ".json": JSONReader,
+    }
+)
+
+
+class IngestionHelper:
+    """Helper class to transform a file into a list of documents.
+
+    This class should be used to transform a file into a list of documents.
+    These methods are thread-safe (and multiprocessing-safe).
+    """
+
+    @staticmethod
+    def transform_file_into_documents(
+        file_name: str, file_data: Path
+    ) -> list[Document]:
+        documents = IngestionHelper._load_file_to_documents(file_name, file_data)
+        for document in documents:
+            document.metadata["file_name"] = file_name
+        IngestionHelper._exclude_metadata(documents)
+        return documents
+
+    @staticmethod
+    def _load_file_to_documents(file_name: str, file_data: Path) -> list[Document]:
+        logger.debug("Transforming file_name=%s into documents", file_name)
+        extension = Path(file_name).suffix
+        reader_cls = FILE_READER_CLS.get(extension)
+        if reader_cls is None:
+            logger.debug(
+                "No reader found for extension=%s, using default string reader",
+                extension,
+            )
+            # Read as a plain text
+            string_reader = StringIterableReader()
+            return string_reader.load_data([file_data.read_text()])
+
+        logger.debug("Specific reader found for extension=%s", extension)
+        return reader_cls().load_data(file_data)
+
+    @staticmethod
+    def _exclude_metadata(documents: list[Document]) -> None:
+        logger.debug("Excluding metadata from count=%s documents", len(documents))
+        for document in documents:
+            document.metadata["doc_id"] = document.doc_id
+            # We don't want the Embeddings search to receive this metadata
+            document.excluded_embed_metadata_keys = ["doc_id"]
+            # We don't want the LLM to receive these metadata in the context
+            document.excluded_llm_metadata_keys = ["file_name", "doc_id", "page_label"]

private_gpt/components/llm/__init__.py

@@ -0,0 +1 @@
+"""LLM implementations."""

private_gpt/components/llm/custom/sagemaker.py

@@ -0,0 +1,276 @@
+# mypy: ignore-errors
+from __future__ import annotations
+
+import io
+import json
+import logging
+from typing import TYPE_CHECKING, Any
+
+import boto3  # type: ignore
+from llama_index.core.base.llms.generic_utils import (
+    completion_response_to_chat_response,
+    stream_completion_response_to_chat_response,
+)
+from llama_index.core.bridge.pydantic import Field
+from llama_index.core.llms import (
+    CompletionResponse,
+    CustomLLM,
+    LLMMetadata,
+)
+from llama_index.core.llms.callbacks import (
+    llm_chat_callback,
+    llm_completion_callback,
+)
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from llama_index.callbacks import CallbackManager
+    from llama_index.llms import (
+        ChatMessage,
+        ChatResponse,
+        ChatResponseGen,
+        CompletionResponseGen,
+    )
+
+logger = logging.getLogger(__name__)
+
+
+class LineIterator:
+    r"""A helper class for parsing the byte stream input from TGI container.
+
+    The output of the model will be in the following format:
+    ```
+    b'data:{"token": {"text": " a"}}\n\n'
+    b'data:{"token": {"text": " challenging"}}\n\n'
+    b'data:{"token": {"text": " problem"
+    b'}}'
+    ...
+    ```
+
+    While usually each PayloadPart event from the event stream will contain a byte array
+    with a full json, this is not guaranteed and some of the json objects may be split
+    across PayloadPart events. For example:
+    ```
+    {'PayloadPart': {'Bytes': b'{"outputs": '}}
+    {'PayloadPart': {'Bytes': b'[" problem"]}\n'}}
+    ```
+
+
+    This class accounts for this by concatenating bytes written via the 'write' function
+    and then exposing a method which will return lines (ending with a '\n' character)
+    within the buffer via the 'scan_lines' function. It maintains the position of the
+    last read position to ensure that previous bytes are not exposed again. It will
+    also save any pending lines that doe not end with a '\n' to make sure truncations
+    are concatinated
+    """
+
+    def __init__(self, stream: Any) -> None:
+        """Line iterator initializer."""
+        self.byte_iterator = iter(stream)
+        self.buffer = io.BytesIO()
+        self.read_pos = 0
+
+    def __iter__(self) -> Any:
+        """Self iterator."""
+        return self
+
+    def __next__(self) -> Any:
+        """Next element from iterator."""
+        while True:
+            self.buffer.seek(self.read_pos)
+            line = self.buffer.readline()
+            if line and line[-1] == ord("\n"):
+                self.read_pos += len(line)
+                return line[:-1]
+            try:
+                chunk = next(self.byte_iterator)
+            except StopIteration:
+                if self.read_pos < self.buffer.getbuffer().nbytes:
+                    continue
+                raise
+            if "PayloadPart" not in chunk:
+                logger.warning("Unknown event type=%s", chunk)
+                continue
+            self.buffer.seek(0, io.SEEK_END)
+            self.buffer.write(chunk["PayloadPart"]["Bytes"])
+
+
+class SagemakerLLM(CustomLLM):
+    """Sagemaker Inference Endpoint models.
+
+    To use, you must supply the endpoint name from your deployed
+    Sagemaker model & the region where it is deployed.
+
+    To authenticate, the AWS client uses the following methods to
+    automatically load credentials:
+    https://boto3.amazonaws.com/v1/documentation/api/latest/guide/credentials.html
+
+    If a specific credential profile should be used, you must pass
+    the name of the profile from the ~/.aws/credentials file that is to be used.
+
+    Make sure the credentials / roles used have the required policies to
+    access the Sagemaker endpoint.
+    See: https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies.html
+    """
+
+    endpoint_name: str = Field(description="")
+    temperature: float = Field(description="The temperature to use for sampling.")
+    max_new_tokens: int = Field(description="The maximum number of tokens to generate.")
+    context_window: int = Field(
+        description="The maximum number of context tokens for the model."
+    )
+    messages_to_prompt: Any = Field(
+        description="The function to convert messages to a prompt.", exclude=True
+    )
+    completion_to_prompt: Any = Field(
+        description="The function to convert a completion to a prompt.", exclude=True
+    )
+    generate_kwargs: dict[str, Any] = Field(
+        default_factory=dict, description="Kwargs used for generation."
+    )
+    model_kwargs: dict[str, Any] = Field(
+        default_factory=dict, description="Kwargs used for model initialization."
+    )
+    verbose: bool = Field(description="Whether to print verbose output.")
+
+    _boto_client: Any = boto3.client(
+        "sagemaker-runtime",
+    )  # TODO make it an optional field
+
+    def __init__(
+        self,
+        endpoint_name: str | None = "",
+        temperature: float = 0.1,
+        max_new_tokens: int = 512,  # to review defaults
+        context_window: int = 2048,  # to review defaults
+        messages_to_prompt: Any = None,
+        completion_to_prompt: Any = None,
+        callback_manager: CallbackManager | None = None,
+        generate_kwargs: dict[str, Any] | None = None,
+        model_kwargs: dict[str, Any] | None = None,
+        verbose: bool = True,
+    ) -> None:
+        """SagemakerLLM initializer."""
+        model_kwargs = model_kwargs or {}
+        model_kwargs.update({"n_ctx": context_window, "verbose": verbose})
+
+        messages_to_prompt = messages_to_prompt or {}
+        completion_to_prompt = completion_to_prompt or {}
+
+        generate_kwargs = generate_kwargs or {}
+        generate_kwargs.update(
+            {"temperature": temperature, "max_tokens": max_new_tokens}
+        )
+
+        super().__init__(
+            endpoint_name=endpoint_name,
+            temperature=temperature,
+            context_window=context_window,
+            max_new_tokens=max_new_tokens,
+            messages_to_prompt=messages_to_prompt,
+            completion_to_prompt=completion_to_prompt,
+            callback_manager=callback_manager,
+            generate_kwargs=generate_kwargs,
+            model_kwargs=model_kwargs,
+            verbose=verbose,
+        )
+
+    @property
+    def inference_params(self):
+        # TODO expose the rest of params
+        return {
+            "do_sample": True,
+            "top_p": 0.7,
+            "temperature": self.temperature,
+            "top_k": 50,
+            "max_new_tokens": self.max_new_tokens,
+        }
+
+    @property
+    def metadata(self) -> LLMMetadata:
+        """Get LLM metadata."""
+        return LLMMetadata(
+            context_window=self.context_window,
+            num_output=self.max_new_tokens,
+            model_name="Sagemaker LLama 2",
+        )
+
+    @llm_completion_callback()
+    def complete(self, prompt: str, **kwargs: Any) -> CompletionResponse:
+        self.generate_kwargs.update({"stream": False})
+
+        is_formatted = kwargs.pop("formatted", False)
+        if not is_formatted:
+            prompt = self.completion_to_prompt(prompt)
+
+        request_params = {
+            "inputs": prompt,
+            "stream": False,
+            "parameters": self.inference_params,
+        }
+
+        resp = self._boto_client.invoke_endpoint(
+            EndpointName=self.endpoint_name,
+            Body=json.dumps(request_params),
+            ContentType="application/json",
+        )
+
+        response_body = resp["Body"]
+        response_str = response_body.read().decode("utf-8")
+        response_dict = json.loads(response_str)
+
+        return CompletionResponse(
+            text=response_dict[0]["generated_text"][len(prompt) :], raw=resp
+        )
+
+    @llm_completion_callback()
+    def stream_complete(self, prompt: str, **kwargs: Any) -> CompletionResponseGen:
+        def get_stream():
+            text = ""
+
+            request_params = {
+                "inputs": prompt,
+                "stream": True,
+                "parameters": self.inference_params,
+            }
+            resp = self._boto_client.invoke_endpoint_with_response_stream(
+                EndpointName=self.endpoint_name,
+                Body=json.dumps(request_params),
+                ContentType="application/json",
+            )
+
+            event_stream = resp["Body"]
+            start_json = b"{"
+            stop_token = "<|endoftext|>"
+            first_token = True
+
+            for line in LineIterator(event_stream):
+                if line != b"" and start_json in line:
+                    data = json.loads(line[line.find(start_json) :].decode("utf-8"))
+                    special = data["token"]["special"]
+                    stop = data["token"]["text"] == stop_token
+                    if not special and not stop:
+                        delta = data["token"]["text"]
+                        # trim the leading space for the first token if present
+                        if first_token:
+                            delta = delta.lstrip()
+                            first_token = False
+                        text += delta
+                        yield CompletionResponse(delta=delta, text=text, raw=data)
+
+        return get_stream()
+
+    @llm_chat_callback()
+    def chat(self, messages: Sequence[ChatMessage], **kwargs: Any) -> ChatResponse:
+        prompt = self.messages_to_prompt(messages)
+        completion_response = self.complete(prompt, formatted=True, **kwargs)
+        return completion_response_to_chat_response(completion_response)
+
+    @llm_chat_callback()
+    def stream_chat(
+        self, messages: Sequence[ChatMessage], **kwargs: Any
+    ) -> ChatResponseGen:
+        prompt = self.messages_to_prompt(messages)
+        completion_response = self.stream_complete(prompt, formatted=True, **kwargs)
+        return stream_completion_response_to_chat_response(completion_response)

private_gpt/components/llm/llm_component.py

@@ -0,0 +1,226 @@
+import logging
+from collections.abc import Callable
+from typing import Any
+
+from injector import inject, singleton
+from llama_index.core.llms import LLM, MockLLM
+from llama_index.core.settings import Settings as LlamaIndexSettings
+from llama_index.core.utils import set_global_tokenizer
+from transformers import AutoTokenizer  # type: ignore
+
+from private_gpt.components.llm.prompt_helper import get_prompt_style
+from private_gpt.paths import models_cache_path, models_path
+from private_gpt.settings.settings import Settings
+
+logger = logging.getLogger(__name__)
+
+
+@singleton
+class LLMComponent:
+    llm: LLM
+
+    @inject
+    def __init__(self, settings: Settings) -> None:
+        llm_mode = settings.llm.mode
+        if settings.llm.tokenizer and settings.llm.mode != "mock":
+            # Try to download the tokenizer. If it fails, the LLM will still work
+            # using the default one, which is less accurate.
+            try:
+                set_global_tokenizer(
+                    AutoTokenizer.from_pretrained(
+                        pretrained_model_name_or_path=settings.llm.tokenizer,
+                        cache_dir=str(models_cache_path),
+                        token=settings.huggingface.access_token,
+                    )
+                )
+            except Exception as e:
+                logger.warning(
+                    f"Failed to download tokenizer {settings.llm.tokenizer}: {e!s}"
+                    f"Please follow the instructions in the documentation to download it if needed: "
+                    f"https://docs.privategpt.dev/installation/getting-started/troubleshooting#tokenizer-setup."
+                    f"Falling back to default tokenizer."
+                )
+
+        logger.info("Initializing the LLM in mode=%s", llm_mode)
+        match settings.llm.mode:
+            case "llamacpp":
+                try:
+                    from llama_index.llms.llama_cpp import LlamaCPP  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "Local dependencies not found, install with `poetry install --extras llms-llama-cpp`"
+                    ) from e
+
+                prompt_style = get_prompt_style(settings.llm.prompt_style)
+                settings_kwargs = {
+                    "tfs_z": settings.llamacpp.tfs_z,  # ollama and llama-cpp
+                    "top_k": settings.llamacpp.top_k,  # ollama and llama-cpp
+                    "top_p": settings.llamacpp.top_p,  # ollama and llama-cpp
+                    "repeat_penalty": settings.llamacpp.repeat_penalty,  # ollama llama-cpp
+                    "n_gpu_layers": -1,
+                    "offload_kqv": True,
+                }
+                self.llm = LlamaCPP(
+                    model_path=str(models_path / settings.llamacpp.llm_hf_model_file),
+                    temperature=settings.llm.temperature,
+                    max_new_tokens=settings.llm.max_new_tokens,
+                    context_window=settings.llm.context_window,
+                    generate_kwargs={},
+                    callback_manager=LlamaIndexSettings.callback_manager,
+                    # All to GPU
+                    model_kwargs=settings_kwargs,
+                    # transform inputs into Llama2 format
+                    messages_to_prompt=prompt_style.messages_to_prompt,
+                    completion_to_prompt=prompt_style.completion_to_prompt,
+                    verbose=True,
+                )
+
+            case "sagemaker":
+                try:
+                    from private_gpt.components.llm.custom.sagemaker import SagemakerLLM
+                except ImportError as e:
+                    raise ImportError(
+                        "Sagemaker dependencies not found, install with `poetry install --extras llms-sagemaker`"
+                    ) from e
+
+                self.llm = SagemakerLLM(
+                    endpoint_name=settings.sagemaker.llm_endpoint_name,
+                    max_new_tokens=settings.llm.max_new_tokens,
+                    context_window=settings.llm.context_window,
+                )
+            case "openai":
+                try:
+                    from llama_index.llms.openai import OpenAI  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "OpenAI dependencies not found, install with `poetry install --extras llms-openai`"
+                    ) from e
+
+                openai_settings = settings.openai
+                self.llm = OpenAI(
+                    api_base=openai_settings.api_base,
+                    api_key=openai_settings.api_key,
+                    model=openai_settings.model,
+                )
+            case "openailike":
+                try:
+                    from llama_index.llms.openai_like import OpenAILike  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "OpenAILike dependencies not found, install with `poetry install --extras llms-openai-like`"
+                    ) from e
+                prompt_style = get_prompt_style(settings.llm.prompt_style)
+                openai_settings = settings.openai
+                self.llm = OpenAILike(
+                    api_base=openai_settings.api_base,
+                    api_key=openai_settings.api_key,
+                    model=openai_settings.model,
+                    is_chat_model=True,
+                    max_tokens=settings.llm.max_new_tokens,
+                    api_version="",
+                    temperature=settings.llm.temperature,
+                    context_window=settings.llm.context_window,
+                    max_new_tokens=settings.llm.max_new_tokens,
+                    messages_to_prompt=prompt_style.messages_to_prompt,
+                    completion_to_prompt=prompt_style.completion_to_prompt,
+                    tokenizer=settings.llm.tokenizer,
+                    timeout=openai_settings.request_timeout,
+                    reuse_client=False,
+                )
+            case "ollama":
+                try:
+                    from llama_index.llms.ollama import Ollama  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "Ollama dependencies not found, install with `poetry install --extras llms-ollama`"
+                    ) from e
+
+                ollama_settings = settings.ollama
+
+                settings_kwargs = {
+                    "tfs_z": ollama_settings.tfs_z,  # ollama and llama-cpp
+                    "num_predict": ollama_settings.num_predict,  # ollama only
+                    "top_k": ollama_settings.top_k,  # ollama and llama-cpp
+                    "top_p": ollama_settings.top_p,  # ollama and llama-cpp
+                    "repeat_last_n": ollama_settings.repeat_last_n,  # ollama
+                    "repeat_penalty": ollama_settings.repeat_penalty,  # ollama llama-cpp
+                }
+
+                # calculate llm model. If not provided tag, it will be use latest
+                model_name = (
+                    ollama_settings.llm_model + ":latest"
+                    if ":" not in ollama_settings.llm_model
+                    else ollama_settings.llm_model
+                )
+
+                llm = Ollama(
+                    model=model_name,
+                    base_url=ollama_settings.api_base,
+                    temperature=settings.llm.temperature,
+                    context_window=settings.llm.context_window,
+                    additional_kwargs=settings_kwargs,
+                    request_timeout=ollama_settings.request_timeout,
+                )
+
+                if ollama_settings.autopull_models:
+                    from private_gpt.utils.ollama import check_connection, pull_model
+
+                    if not check_connection(llm.client):
+                        raise ValueError(
+                            f"Failed to connect to Ollama, "
+                            f"check if Ollama server is running on {ollama_settings.api_base}"
+                        )
+                    pull_model(llm.client, model_name)
+
+                if (
+                    ollama_settings.keep_alive
+                    != ollama_settings.model_fields["keep_alive"].default
+                ):
+                    # Modify Ollama methods to use the "keep_alive" field.
+                    def add_keep_alive(func: Callable[..., Any]) -> Callable[..., Any]:
+                        def wrapper(*args: Any, **kwargs: Any) -> Any:
+                            kwargs["keep_alive"] = ollama_settings.keep_alive
+                            return func(*args, **kwargs)
+
+                        return wrapper
+
+                    Ollama.chat = add_keep_alive(Ollama.chat)
+                    Ollama.stream_chat = add_keep_alive(Ollama.stream_chat)
+                    Ollama.complete = add_keep_alive(Ollama.complete)
+                    Ollama.stream_complete = add_keep_alive(Ollama.stream_complete)
+
+                self.llm = llm
+
+            case "azopenai":
+                try:
+                    from llama_index.llms.azure_openai import (  # type: ignore
+                        AzureOpenAI,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Azure OpenAI dependencies not found, install with `poetry install --extras llms-azopenai`"
+                    ) from e
+
+                azopenai_settings = settings.azopenai
+                self.llm = AzureOpenAI(
+                    model=azopenai_settings.llm_model,
+                    deployment_name=azopenai_settings.llm_deployment_name,
+                    api_key=azopenai_settings.api_key,
+                    azure_endpoint=azopenai_settings.azure_endpoint,
+                    api_version=azopenai_settings.api_version,
+                )
+            case "gemini":
+                try:
+                    from llama_index.llms.gemini import (  # type: ignore
+                        Gemini,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Google Gemini dependencies not found, install with `poetry install --extras llms-gemini`"
+                    ) from e
+                gemini_settings = settings.gemini
+                self.llm = Gemini(
+                    model_name=gemini_settings.model, api_key=gemini_settings.api_key
+                )
+            case "mock":
+                self.llm = MockLLM()

private_gpt/components/llm/prompt_helper.py

@@ -0,0 +1,308 @@
+import abc
+import logging
+from collections.abc import Sequence
+from typing import Any, Literal
+
+from llama_index.core.llms import ChatMessage, MessageRole
+
+logger = logging.getLogger(__name__)
+
+
+class AbstractPromptStyle(abc.ABC):
+    """Abstract class for prompt styles.
+
+    This class is used to format a series of messages into a prompt that can be
+    understood by the models. A series of messages represents the interaction(s)
+    between a user and an assistant. This series of messages can be considered as a
+    session between a user X and an assistant Y.This session holds, through the
+    messages, the state of the conversation. This session, to be understood by the
+    model, needs to be formatted into a prompt (i.e. a string that the models
+    can understand). Prompts can be formatted in different ways,
+    depending on the model.
+
+    The implementations of this class represent the different ways to format a
+    series of messages into a prompt.
+    """
+
+    def __init__(self, *args: Any, **kwargs: Any) -> None:
+        logger.debug("Initializing prompt_style=%s", self.__class__.__name__)
+
+    @abc.abstractmethod
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        pass
+
+    @abc.abstractmethod
+    def _completion_to_prompt(self, completion: str) -> str:
+        pass
+
+    def messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        prompt = self._messages_to_prompt(messages)
+        logger.debug("Got for messages='%s' the prompt='%s'", messages, prompt)
+        return prompt
+
+    def completion_to_prompt(self, completion: str) -> str:
+        prompt = self._completion_to_prompt(completion)
+        logger.debug("Got for completion='%s' the prompt='%s'", completion, prompt)
+        return prompt
+
+
+class DefaultPromptStyle(AbstractPromptStyle):
+    """Default prompt style that uses the defaults from llama_utils.
+
+    It basically passes None to the LLM, indicating it should use
+    the default functions.
+    """
+
+    def __init__(self, *args: Any, **kwargs: Any) -> None:
+        super().__init__(*args, **kwargs)
+
+        # Hacky way to override the functions
+        # Override the functions to be None, and pass None to the LLM.
+        self.messages_to_prompt = None  # type: ignore[method-assign, assignment]
+        self.completion_to_prompt = None  # type: ignore[method-assign, assignment]
+
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        return ""
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        return ""
+
+
+class Llama2PromptStyle(AbstractPromptStyle):
+    """Simple prompt style that uses llama 2 prompt style.
+
+    Inspired by llama_index/legacy/llms/llama_utils.py
+
+    It transforms the sequence of messages into a prompt that should look like:
+    ```text
+    <s> [INST] <<SYS>> your system prompt here. <</SYS>>
+
+    user message here [/INST] assistant (model) response here </s>
+    ```
+    """
+
+    BOS, EOS = "<s>", "</s>"
+    B_INST, E_INST = "[INST]", "[/INST]"
+    B_SYS, E_SYS = "<<SYS>>\n", "\n<</SYS>>\n\n"
+    DEFAULT_SYSTEM_PROMPT = """\
+    You are a helpful, respectful and honest assistant. \
+    Always answer as helpfully as possible and follow ALL given instructions. \
+    Do not speculate or make up information. \
+    Do not reference any given instructions or context. \
+    """
+
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        string_messages: list[str] = []
+        if messages[0].role == MessageRole.SYSTEM:
+            # pull out the system message (if it exists in messages)
+            system_message_str = messages[0].content or ""
+            messages = messages[1:]
+        else:
+            system_message_str = self.DEFAULT_SYSTEM_PROMPT
+
+        system_message_str = f"{self.B_SYS} {system_message_str.strip()} {self.E_SYS}"
+
+        for i in range(0, len(messages), 2):
+            # first message should always be a user
+            user_message = messages[i]
+            assert user_message.role == MessageRole.USER
+
+            if i == 0:
+                # make sure system prompt is included at the start
+                str_message = f"{self.BOS} {self.B_INST} {system_message_str} "
+            else:
+                # end previous user-assistant interaction
+                string_messages[-1] += f" {self.EOS}"
+                # no need to include system prompt
+                str_message = f"{self.BOS} {self.B_INST} "
+
+            # include user message content
+            str_message += f"{user_message.content} {self.E_INST}"
+
+            if len(messages) > (i + 1):
+                # if assistant message exists, add to str_message
+                assistant_message = messages[i + 1]
+                assert assistant_message.role == MessageRole.ASSISTANT
+                str_message += f" {assistant_message.content}"
+
+            string_messages.append(str_message)
+
+        return "".join(string_messages)
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        system_prompt_str = self.DEFAULT_SYSTEM_PROMPT
+
+        return (
+            f"{self.BOS} {self.B_INST} {self.B_SYS} {system_prompt_str.strip()} {self.E_SYS} "
+            f"{completion.strip()} {self.E_INST}"
+        )
+
+
+class Llama3PromptStyle(AbstractPromptStyle):
+    r"""Template for Meta's Llama 3.1.
+
+    The format follows this structure:
+    <|begin_of_text|>
+    <|start_header_id|>system<|end_header_id|>
+
+    [System message content]<|eot_id|>
+    <|start_header_id|>user<|end_header_id|>
+
+    [User message content]<|eot_id|>
+    <|start_header_id|>assistant<|end_header_id|>
+
+    [Assistant message content]<|eot_id|>
+    ...
+    (Repeat for each message, including possible 'ipython' role)
+    """
+
+    BOS, EOS = "<|begin_of_text|>", "<|end_of_text|>"
+    B_INST, E_INST = "<|start_header_id|>", "<|end_header_id|>"
+    EOT = "<|eot_id|>"
+    B_SYS, E_SYS = "<|start_header_id|>system<|end_header_id|>", "<|eot_id|>"
+    ASSISTANT_INST = "<|start_header_id|>assistant<|end_header_id|>"
+    DEFAULT_SYSTEM_PROMPT = """\
+    You are a helpful, respectful and honest assistant. \
+    Always answer as helpfully as possible and follow ALL given instructions. \
+    Do not speculate or make up information. \
+    Do not reference any given instructions or context. \
+    """
+
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        prompt = ""
+        has_system_message = False
+
+        for i, message in enumerate(messages):
+            if not message or message.content is None:
+                continue
+            if message.role == MessageRole.SYSTEM:
+                prompt += f"{self.B_SYS}\n\n{message.content.strip()}{self.E_SYS}"
+                has_system_message = True
+            else:
+                role_header = f"{self.B_INST}{message.role.value}{self.E_INST}"
+                prompt += f"{role_header}\n\n{message.content.strip()}{self.EOT}"
+
+            # Add assistant header if the last message is not from the assistant
+            if i == len(messages) - 1 and message.role != MessageRole.ASSISTANT:
+                prompt += f"{self.ASSISTANT_INST}\n\n"
+
+        # Add default system prompt if no system message was provided
+        if not has_system_message:
+            prompt = (
+                f"{self.B_SYS}\n\n{self.DEFAULT_SYSTEM_PROMPT}{self.E_SYS}" + prompt
+            )
+
+        # TODO: Implement tool handling logic
+
+        return prompt
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        return (
+            f"{self.B_SYS}\n\n{self.DEFAULT_SYSTEM_PROMPT}{self.E_SYS}"
+            f"{self.B_INST}user{self.E_INST}\n\n{completion.strip()}{self.EOT}"
+            f"{self.ASSISTANT_INST}\n\n"
+        )
+
+
+class TagPromptStyle(AbstractPromptStyle):
+    """Tag prompt style (used by Vigogne) that uses the prompt style `<|ROLE|>`.
+
+    It transforms the sequence of messages into a prompt that should look like:
+    ```text
+    <|system|>: your system prompt here.
+    <|user|>: user message here
+    (possibly with context and question)
+    <|assistant|>: assistant (model) response here.
+    ```
+
+    FIXME: should we add surrounding `<s>` and `</s>` tags, like in llama2?
+    """
+
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        """Format message to prompt with `<|ROLE|>: MSG` style."""
+        prompt = ""
+        for message in messages:
+            role = message.role
+            content = message.content or ""
+            message_from_user = f"<|{role.lower()}|>: {content.strip()}"
+            message_from_user += "\n"
+            prompt += message_from_user
+        # we are missing the last <|assistant|> tag that will trigger a completion
+        prompt += "<|assistant|>: "
+        return prompt
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        return self._messages_to_prompt(
+            [ChatMessage(content=completion, role=MessageRole.USER)]
+        )
+
+
+class MistralPromptStyle(AbstractPromptStyle):
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        inst_buffer = []
+        text = ""
+        for message in messages:
+            if message.role == MessageRole.SYSTEM or message.role == MessageRole.USER:
+                inst_buffer.append(str(message.content).strip())
+            elif message.role == MessageRole.ASSISTANT:
+                text += "<s>[INST] " + "\n".join(inst_buffer) + " [/INST]"
+                text += " " + str(message.content).strip() + "</s>"
+                inst_buffer.clear()
+            else:
+                raise ValueError(f"Unknown message role {message.role}")
+
+        if len(inst_buffer) > 0:
+            text += "<s>[INST] " + "\n".join(inst_buffer) + " [/INST]"
+
+        return text
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        return self._messages_to_prompt(
+            [ChatMessage(content=completion, role=MessageRole.USER)]
+        )
+
+
+class ChatMLPromptStyle(AbstractPromptStyle):
+    def _messages_to_prompt(self, messages: Sequence[ChatMessage]) -> str:
+        prompt = "<|im_start|>system\n"
+        for message in messages:
+            role = message.role
+            content = message.content or ""
+            if role.lower() == "system":
+                message_from_user = f"{content.strip()}"
+                prompt += message_from_user
+            elif role.lower() == "user":
+                prompt += "<|im_end|>\n<|im_start|>user\n"
+                message_from_user = f"{content.strip()}<|im_end|>\n"
+                prompt += message_from_user
+        prompt += "<|im_start|>assistant\n"
+        return prompt
+
+    def _completion_to_prompt(self, completion: str) -> str:
+        return self._messages_to_prompt(
+            [ChatMessage(content=completion, role=MessageRole.USER)]
+        )
+
+
+def get_prompt_style(
+    prompt_style: Literal["default", "llama2", "llama3", "tag", "mistral", "chatml"]
+    | None
+) -> AbstractPromptStyle:
+    """Get the prompt style to use from the given string.
+
+    :param prompt_style: The prompt style to use.
+    :return: The prompt style to use.
+    """
+    if prompt_style is None or prompt_style == "default":
+        return DefaultPromptStyle()
+    elif prompt_style == "llama2":
+        return Llama2PromptStyle()
+    elif prompt_style == "llama3":
+        return Llama3PromptStyle()
+    elif prompt_style == "tag":
+        return TagPromptStyle()
+    elif prompt_style == "mistral":
+        return MistralPromptStyle()
+    elif prompt_style == "chatml":
+        return ChatMLPromptStyle()
+    raise ValueError(f"Unknown prompt_style='{prompt_style}'")

private_gpt/components/node_store/node_store_component.py

@@ -0,0 +1,67 @@
+import logging
+
+from injector import inject, singleton
+from llama_index.core.storage.docstore import BaseDocumentStore, SimpleDocumentStore
+from llama_index.core.storage.index_store import SimpleIndexStore
+from llama_index.core.storage.index_store.types import BaseIndexStore
+
+from private_gpt.paths import local_data_path
+from private_gpt.settings.settings import Settings
+
+logger = logging.getLogger(__name__)
+
+
+@singleton
+class NodeStoreComponent:
+    index_store: BaseIndexStore
+    doc_store: BaseDocumentStore
+
+    @inject
+    def __init__(self, settings: Settings) -> None:
+        match settings.nodestore.database:
+            case "simple":
+                try:
+                    self.index_store = SimpleIndexStore.from_persist_dir(
+                        persist_dir=str(local_data_path)
+                    )
+                except FileNotFoundError:
+                    logger.debug("Local index store not found, creating a new one")
+                    self.index_store = SimpleIndexStore()
+
+                try:
+                    self.doc_store = SimpleDocumentStore.from_persist_dir(
+                        persist_dir=str(local_data_path)
+                    )
+                except FileNotFoundError:
+                    logger.debug("Local document store not found, creating a new one")
+                    self.doc_store = SimpleDocumentStore()
+
+            case "postgres":
+                try:
+                    from llama_index.core.storage.docstore.postgres_docstore import (
+                        PostgresDocumentStore,
+                    )
+                    from llama_index.core.storage.index_store.postgres_index_store import (
+                        PostgresIndexStore,
+                    )
+                except ImportError:
+                    raise ImportError(
+                        "Postgres dependencies not found, install with `poetry install --extras storage-nodestore-postgres`"
+                    ) from None
+
+                if settings.postgres is None:
+                    raise ValueError("Postgres index/doc store settings not found.")
+
+                self.index_store = PostgresIndexStore.from_params(
+                    **settings.postgres.model_dump(exclude_none=True)
+                )
+                self.doc_store = PostgresDocumentStore.from_params(
+                    **settings.postgres.model_dump(exclude_none=True)
+                )
+
+            case _:
+                # Should be unreachable
+                # The settings validator should have caught this
+                raise ValueError(
+                    f"Database {settings.nodestore.database} not supported"
+                )

private_gpt/components/vector_store/batched_chroma.py

@@ -0,0 +1,103 @@
+from collections.abc import Generator
+from typing import Any
+
+from llama_index.core.schema import BaseNode, MetadataMode
+from llama_index.core.vector_stores.utils import node_to_metadata_dict
+from llama_index.vector_stores.chroma import ChromaVectorStore  # type: ignore
+
+
+def chunk_list(
+    lst: list[BaseNode], max_chunk_size: int
+) -> Generator[list[BaseNode], None, None]:
+    """Yield successive max_chunk_size-sized chunks from lst.
+
+    Args:
+        lst (List[BaseNode]): list of nodes with embeddings
+        max_chunk_size (int): max chunk size
+
+    Yields:
+        Generator[List[BaseNode], None, None]: list of nodes with embeddings
+    """
+    for i in range(0, len(lst), max_chunk_size):
+        yield lst[i : i + max_chunk_size]
+
+
+class BatchedChromaVectorStore(ChromaVectorStore):  # type: ignore
+    """Chroma vector store, batching additions to avoid reaching the max batch limit.
+
+    In this vector store, embeddings are stored within a ChromaDB collection.
+
+    During query time, the index uses ChromaDB to query for the top
+    k most similar nodes.
+
+    Args:
+        chroma_client (from chromadb.api.API):
+            API instance
+        chroma_collection (chromadb.api.models.Collection.Collection):
+            ChromaDB collection instance
+
+    """
+
+    chroma_client: Any | None
+
+    def __init__(
+        self,
+        chroma_client: Any,
+        chroma_collection: Any,
+        host: str | None = None,
+        port: str | None = None,
+        ssl: bool = False,
+        headers: dict[str, str] | None = None,
+        collection_kwargs: dict[Any, Any] | None = None,
+    ) -> None:
+        super().__init__(
+            chroma_collection=chroma_collection,
+            host=host,
+            port=port,
+            ssl=ssl,
+            headers=headers,
+            collection_kwargs=collection_kwargs or {},
+        )
+        self.chroma_client = chroma_client
+
+    def add(self, nodes: list[BaseNode], **add_kwargs: Any) -> list[str]:
+        """Add nodes to index, batching the insertion to avoid issues.
+
+        Args:
+            nodes: List[BaseNode]: list of nodes with embeddings
+            add_kwargs: _
+        """
+        if not self.chroma_client:
+            raise ValueError("Client not initialized")
+
+        if not self._collection:
+            raise ValueError("Collection not initialized")
+
+        max_chunk_size = self.chroma_client.max_batch_size
+        node_chunks = chunk_list(nodes, max_chunk_size)
+
+        all_ids = []
+        for node_chunk in node_chunks:
+            embeddings = []
+            metadatas = []
+            ids = []
+            documents = []
+            for node in node_chunk:
+                embeddings.append(node.get_embedding())
+                metadatas.append(
+                    node_to_metadata_dict(
+                        node, remove_text=True, flat_metadata=self.flat_metadata
+                    )
+                )
+                ids.append(node.node_id)
+                documents.append(node.get_content(metadata_mode=MetadataMode.NONE))
+
+            self._collection.add(
+                embeddings=embeddings,
+                ids=ids,
+                metadatas=metadatas,
+                documents=documents,
+            )
+            all_ids.extend(ids)
+
+        return all_ids

private_gpt/components/vector_store/vector_store_component.py

@@ -0,0 +1,217 @@
+import logging
+import typing
+
+from injector import inject, singleton
+from llama_index.core.indices.vector_store import VectorIndexRetriever, VectorStoreIndex
+from llama_index.core.vector_stores.types import (
+    BasePydanticVectorStore,
+    FilterCondition,
+    MetadataFilter,
+    MetadataFilters,
+)
+
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.paths import local_data_path
+from private_gpt.settings.settings import Settings
+
+logger = logging.getLogger(__name__)
+
+
+def _doc_id_metadata_filter(
+    context_filter: ContextFilter | None,
+) -> MetadataFilters:
+    filters = MetadataFilters(filters=[], condition=FilterCondition.OR)
+
+    if context_filter is not None and context_filter.docs_ids is not None:
+        for doc_id in context_filter.docs_ids:
+            filters.filters.append(MetadataFilter(key="doc_id", value=doc_id))
+
+    return filters
+
+
+@singleton
+class VectorStoreComponent:
+    settings: Settings
+    vector_store: BasePydanticVectorStore
+
+    @inject
+    def __init__(self, settings: Settings) -> None:
+        self.settings = settings
+        match settings.vectorstore.database:
+            case "postgres":
+                try:
+                    from llama_index.vector_stores.postgres import (  # type: ignore
+                        PGVectorStore,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Postgres dependencies not found, install with `poetry install --extras vector-stores-postgres`"
+                    ) from e
+
+                if settings.postgres is None:
+                    raise ValueError(
+                        "Postgres settings not found. Please provide settings."
+                    )
+
+                self.vector_store = typing.cast(
+                    BasePydanticVectorStore,
+                    PGVectorStore.from_params(
+                        **settings.postgres.model_dump(exclude_none=True),
+                        table_name="embeddings",
+                        embed_dim=settings.embedding.embed_dim,
+                    ),
+                )
+
+            case "chroma":
+                try:
+                    import chromadb  # type: ignore
+                    from chromadb.config import (  # type: ignore
+                        Settings as ChromaSettings,
+                    )
+
+                    from private_gpt.components.vector_store.batched_chroma import (
+                        BatchedChromaVectorStore,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "ChromaDB dependencies not found, install with `poetry install --extras vector-stores-chroma`"
+                    ) from e
+
+                chroma_settings = ChromaSettings(anonymized_telemetry=False)
+                chroma_client = chromadb.PersistentClient(
+                    path=str((local_data_path / "chroma_db").absolute()),
+                    settings=chroma_settings,
+                )
+                chroma_collection = chroma_client.get_or_create_collection(
+                    "make_this_parameterizable_per_api_call"
+                )  # TODO
+
+                self.vector_store = typing.cast(
+                    BasePydanticVectorStore,
+                    BatchedChromaVectorStore(
+                        chroma_client=chroma_client, chroma_collection=chroma_collection
+                    ),
+                )
+
+            case "qdrant":
+                try:
+                    from llama_index.vector_stores.qdrant import (  # type: ignore
+                        QdrantVectorStore,
+                    )
+                    from qdrant_client import QdrantClient  # type: ignore
+                except ImportError as e:
+                    raise ImportError(
+                        "Qdrant dependencies not found, install with `poetry install --extras vector-stores-qdrant`"
+                    ) from e
+
+                if settings.qdrant is None:
+                    logger.info(
+                        "Qdrant config not found. Using default settings."
+                        "Trying to connect to Qdrant at localhost:6333."
+                    )
+                    client = QdrantClient()
+                else:
+                    client = QdrantClient(
+                        **settings.qdrant.model_dump(exclude_none=True)
+                    )
+                self.vector_store = typing.cast(
+                    BasePydanticVectorStore,
+                    QdrantVectorStore(
+                        client=client,
+                        collection_name="make_this_parameterizable_per_api_call",
+                    ),  # TODO
+                )
+
+            case "milvus":
+                try:
+                    from llama_index.vector_stores.milvus import (  # type: ignore
+                        MilvusVectorStore,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "Milvus dependencies not found, install with `poetry install --extras vector-stores-milvus`"
+                    ) from e
+
+                if settings.milvus is None:
+                    logger.info(
+                        "Milvus config not found. Using default settings.\n"
+                        "Trying to connect to Milvus at local_data/private_gpt/milvus/milvus_local.db "
+                        "with collection 'make_this_parameterizable_per_api_call'."
+                    )
+
+                    self.vector_store = typing.cast(
+                        BasePydanticVectorStore,
+                        MilvusVectorStore(
+                            dim=settings.embedding.embed_dim,
+                            collection_name="make_this_parameterizable_per_api_call",
+                            overwrite=True,
+                        ),
+                    )
+
+                else:
+                    self.vector_store = typing.cast(
+                        BasePydanticVectorStore,
+                        MilvusVectorStore(
+                            dim=settings.embedding.embed_dim,
+                            uri=settings.milvus.uri,
+                            token=settings.milvus.token,
+                            collection_name=settings.milvus.collection_name,
+                            overwrite=settings.milvus.overwrite,
+                        ),
+                    )
+
+            case "clickhouse":
+                try:
+                    from clickhouse_connect import (  # type: ignore
+                        get_client,
+                    )
+                    from llama_index.vector_stores.clickhouse import (  # type: ignore
+                        ClickHouseVectorStore,
+                    )
+                except ImportError as e:
+                    raise ImportError(
+                        "ClickHouse dependencies not found, install with `poetry install --extras vector-stores-clickhouse`"
+                    ) from e
+
+                if settings.clickhouse is None:
+                    raise ValueError(
+                        "ClickHouse settings not found. Please provide settings."
+                    )
+
+                clickhouse_client = get_client(
+                    host=settings.clickhouse.host,
+                    port=settings.clickhouse.port,
+                    username=settings.clickhouse.username,
+                    password=settings.clickhouse.password,
+                )
+                self.vector_store = ClickHouseVectorStore(
+                    clickhouse_client=clickhouse_client
+                )
+            case _:
+                # Should be unreachable
+                # The settings validator should have caught this
+                raise ValueError(
+                    f"Vectorstore database {settings.vectorstore.database} not supported"
+                )
+
+    def get_retriever(
+        self,
+        index: VectorStoreIndex,
+        context_filter: ContextFilter | None = None,
+        similarity_top_k: int = 2,
+    ) -> VectorIndexRetriever:
+        # This way we support qdrant (using doc_ids) and the rest (using filters)
+        return VectorIndexRetriever(
+            index=index,
+            similarity_top_k=similarity_top_k,
+            doc_ids=context_filter.docs_ids if context_filter else None,
+            filters=(
+                _doc_id_metadata_filter(context_filter)
+                if self.settings.vectorstore.database != "qdrant"
+                else None
+            ),
+        )
+
+    def close(self) -> None:
+        if hasattr(self.vector_store.client, "close"):
+            self.vector_store.client.close()

private_gpt/constants.py

@@ -0,0 +1,3 @@
+from pathlib import Path
+
+PROJECT_ROOT_PATH: Path = Path(__file__).parents[1]

private_gpt/di.py

@@ -0,0 +1,19 @@
+from injector import Injector
+
+from private_gpt.settings.settings import Settings, unsafe_typed_settings
+
+
+def create_application_injector() -> Injector:
+    _injector = Injector(auto_bind=True)
+    _injector.binder.bind(Settings, to=unsafe_typed_settings)
+    return _injector
+
+
+"""
+Global injector for the application.
+
+Avoid using this reference, it will make your code harder to test.
+
+Instead, use the `request.state.injector` reference, which is bound to every request
+"""
+global_injector: Injector = create_application_injector()

private_gpt/launcher.py

@@ -0,0 +1,69 @@
+"""FastAPI app creation, logger configuration and main API routes."""
+
+import logging
+
+from fastapi import Depends, FastAPI, Request
+from fastapi.middleware.cors import CORSMiddleware
+from injector import Injector
+from llama_index.core.callbacks import CallbackManager
+from llama_index.core.callbacks.global_handlers import create_global_handler
+from llama_index.core.settings import Settings as LlamaIndexSettings
+
+from private_gpt.server.chat.chat_router import chat_router
+from private_gpt.server.chunks.chunks_router import chunks_router
+from private_gpt.server.completions.completions_router import completions_router
+from private_gpt.server.embeddings.embeddings_router import embeddings_router
+from private_gpt.server.health.health_router import health_router
+from private_gpt.server.ingest.ingest_router import ingest_router
+from private_gpt.server.recipes.summarize.summarize_router import summarize_router
+from private_gpt.settings.settings import Settings
+
+logger = logging.getLogger(__name__)
+
+
+def create_app(root_injector: Injector) -> FastAPI:
+
+    # Start the API
+    async def bind_injector_to_request(request: Request) -> None:
+        request.state.injector = root_injector
+
+    app = FastAPI(dependencies=[Depends(bind_injector_to_request)])
+
+    app.include_router(completions_router)
+    app.include_router(chat_router)
+    app.include_router(chunks_router)
+    app.include_router(ingest_router)
+    app.include_router(summarize_router)
+    app.include_router(embeddings_router)
+    app.include_router(health_router)
+
+    # Add LlamaIndex simple observability
+    global_handler = create_global_handler("simple")
+    if global_handler:
+        LlamaIndexSettings.callback_manager = CallbackManager([global_handler])
+
+    settings = root_injector.get(Settings)
+    if settings.server.cors.enabled:
+        logger.debug("Setting up CORS middleware")
+        app.add_middleware(
+            CORSMiddleware,
+            allow_credentials=settings.server.cors.allow_credentials,
+            allow_origins=settings.server.cors.allow_origins,
+            allow_origin_regex=settings.server.cors.allow_origin_regex,
+            allow_methods=settings.server.cors.allow_methods,
+            allow_headers=settings.server.cors.allow_headers,
+        )
+
+    if settings.ui.enabled:
+        logger.debug("Importing the UI module")
+        try:
+            from private_gpt.ui.ui import PrivateGptUi
+        except ImportError as e:
+            raise ImportError(
+                "UI dependencies not found, install with `poetry install --extras ui`"
+            ) from e
+
+        ui = root_injector.get(PrivateGptUi)
+        ui.mount_in_app(app, settings.ui.path)
+
+    return app

private_gpt/main.py

@@ -0,0 +1,6 @@
+"""FastAPI app creation, logger configuration and main API routes."""
+
+from private_gpt.di import global_injector
+from private_gpt.launcher import create_app
+
+app = create_app(global_injector)

private_gpt/open_ai/__init__.py

@@ -0,0 +1 @@
+"""OpenAI compatibility utilities."""

private_gpt/open_ai/extensions/__init__.py

@@ -0,0 +1 @@
+"""OpenAI API extensions."""

private_gpt/open_ai/extensions/context_filter.py

@@ -0,0 +1,7 @@
+from pydantic import BaseModel, Field
+
+
+class ContextFilter(BaseModel):
+    docs_ids: list[str] | None = Field(
+        examples=[["c202d5e6-7b69-4869-81cc-dd574ee8ee11"]]
+    )

private_gpt/open_ai/openai_models.py

@@ -0,0 +1,122 @@
+import time
+import uuid
+from collections.abc import Iterator
+from typing import Literal
+
+from llama_index.core.llms import ChatResponse, CompletionResponse
+from pydantic import BaseModel, Field
+
+from private_gpt.server.chunks.chunks_service import Chunk
+
+
+class OpenAIDelta(BaseModel):
+    """A piece of completion that needs to be concatenated to get the full message."""
+
+    content: str | None
+
+
+class OpenAIMessage(BaseModel):
+    """Inference result, with the source of the message.
+
+    Role could be the assistant or system
+    (providing a default response, not AI generated).
+    """
+
+    role: Literal["assistant", "system", "user"] = Field(default="user")
+    content: str | None
+
+
+class OpenAIChoice(BaseModel):
+    """Response from AI.
+
+    Either the delta or the message will be present, but never both.
+    Sources used will be returned in case context retrieval was enabled.
+    """
+
+    finish_reason: str | None = Field(examples=["stop"])
+    delta: OpenAIDelta | None = None
+    message: OpenAIMessage | None = None
+    sources: list[Chunk] | None = None
+    index: int = 0
+
+
+class OpenAICompletion(BaseModel):
+    """Clone of OpenAI Completion model.
+
+    For more information see: https://platform.openai.com/docs/api-reference/chat/object
+    """
+
+    id: str
+    object: Literal["completion", "completion.chunk"] = Field(default="completion")
+    created: int = Field(..., examples=[1623340000])
+    model: Literal["private-gpt"]
+    choices: list[OpenAIChoice]
+
+    @classmethod
+    def from_text(
+        cls,
+        text: str | None,
+        finish_reason: str | None = None,
+        sources: list[Chunk] | None = None,
+    ) -> "OpenAICompletion":
+        return OpenAICompletion(
+            id=str(uuid.uuid4()),
+            object="completion",
+            created=int(time.time()),
+            model="private-gpt",
+            choices=[
+                OpenAIChoice(
+                    message=OpenAIMessage(role="assistant", content=text),
+                    finish_reason=finish_reason,
+                    sources=sources,
+                )
+            ],
+        )
+
+    @classmethod
+    def json_from_delta(
+        cls,
+        *,
+        text: str | None,
+        finish_reason: str | None = None,
+        sources: list[Chunk] | None = None,
+    ) -> str:
+        chunk = OpenAICompletion(
+            id=str(uuid.uuid4()),
+            object="completion.chunk",
+            created=int(time.time()),
+            model="private-gpt",
+            choices=[
+                OpenAIChoice(
+                    delta=OpenAIDelta(content=text),
+                    finish_reason=finish_reason,
+                    sources=sources,
+                )
+            ],
+        )
+
+        return chunk.model_dump_json()
+
+
+def to_openai_response(
+    response: str | ChatResponse, sources: list[Chunk] | None = None
+) -> OpenAICompletion:
+    if isinstance(response, ChatResponse):
+        return OpenAICompletion.from_text(response.delta, finish_reason="stop")
+    else:
+        return OpenAICompletion.from_text(
+            response, finish_reason="stop", sources=sources
+        )
+
+
+def to_openai_sse_stream(
+    response_generator: Iterator[str | CompletionResponse | ChatResponse],
+    sources: list[Chunk] | None = None,
+) -> Iterator[str]:
+    for response in response_generator:
+        if isinstance(response, CompletionResponse | ChatResponse):
+            yield f"data: {OpenAICompletion.json_from_delta(text=response.delta)}\n\n"
+        else:
+            yield f"data: {OpenAICompletion.json_from_delta(text=response, sources=sources)}\n\n"
+    yield f"data: {OpenAICompletion.json_from_delta(text='', finish_reason='stop')}\n\n"
+    yield "data: [DONE]\n\n"

private_gpt/paths.py

@@ -0,0 +1,18 @@
+from pathlib import Path
+
+from private_gpt.constants import PROJECT_ROOT_PATH
+from private_gpt.settings.settings import settings
+
+
+def _absolute_or_from_project_root(path: str) -> Path:
+    if path.startswith("/"):
+        return Path(path)
+    return PROJECT_ROOT_PATH / path
+
+
+models_path: Path = PROJECT_ROOT_PATH / "models"
+models_cache_path: Path = models_path / "cache"
+docs_path: Path = PROJECT_ROOT_PATH / "docs"
+local_data_path: Path = _absolute_or_from_project_root(
+    settings().data.local_data_folder
+)

private_gpt/server/__init__.py

@@ -0,0 +1 @@
+"""private-gpt server."""

private_gpt/server/chat/chat_router.py

@@ -0,0 +1,115 @@
+from fastapi import APIRouter, Depends, Request
+from llama_index.core.llms import ChatMessage, MessageRole
+from pydantic import BaseModel
+from starlette.responses import StreamingResponse
+
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.open_ai.openai_models import (
+    OpenAICompletion,
+    OpenAIMessage,
+    to_openai_response,
+    to_openai_sse_stream,
+)
+from private_gpt.server.chat.chat_service import ChatService
+from private_gpt.server.utils.auth import authenticated
+
+chat_router = APIRouter(prefix="/v1", dependencies=[Depends(authenticated)])
+
+
+class ChatBody(BaseModel):
+    messages: list[OpenAIMessage]
+    use_context: bool = False
+    context_filter: ContextFilter | None = None
+    include_sources: bool = True
+    stream: bool = False
+
+    model_config = {
+        "json_schema_extra": {
+            "examples": [
+                {
+                    "messages": [
+                        {
+                            "role": "system",
+                            "content": "You are a rapper. Always answer with a rap.",
+                        },
+                        {
+                            "role": "user",
+                            "content": "How do you fry an egg?",
+                        },
+                    ],
+                    "stream": False,
+                    "use_context": True,
+                    "include_sources": True,
+                    "context_filter": {
+                        "docs_ids": ["c202d5e6-7b69-4869-81cc-dd574ee8ee11"]
+                    },
+                }
+            ]
+        }
+    }
+
+
+@chat_router.post(
+    "/chat/completions",
+    response_model=None,
+    responses={200: {"model": OpenAICompletion}},
+    tags=["Contextual Completions"],
+    openapi_extra={
+        "x-fern-streaming": {
+            "stream-condition": "stream",
+            "response": {"$ref": "#/components/schemas/OpenAICompletion"},
+            "response-stream": {"$ref": "#/components/schemas/OpenAICompletion"},
+        }
+    },
+)
+def chat_completion(
+    request: Request, body: ChatBody
+) -> OpenAICompletion | StreamingResponse:
+    """Given a list of messages comprising a conversation, return a response.
+
+    Optionally include an initial `role: system` message to influence the way
+    the LLM answers.
+
+    If `use_context` is set to `true`, the model will use context coming
+    from the ingested documents to create the response. The documents being used can
+    be filtered using the `context_filter` and passing the document IDs to be used.
+    Ingested documents IDs can be found using `/ingest/list` endpoint. If you want
+    all ingested documents to be used, remove `context_filter` altogether.
+
+    When using `'include_sources': true`, the API will return the source Chunks used
+    to create the response, which come from the context provided.
+
+    When using `'stream': true`, the API will return data chunks following [OpenAI's
+    streaming model](https://platform.openai.com/docs/api-reference/chat/streaming):
+    ```
+    {"id":"12345","object":"completion.chunk","created":1694268190,
+    "model":"private-gpt","choices":[{"index":0,"delta":{"content":"Hello"},
+    "finish_reason":null}]}
+    ```
+    """
+    service = request.state.injector.get(ChatService)
+    all_messages = [
+        ChatMessage(content=m.content, role=MessageRole(m.role)) for m in body.messages
+    ]
+    if body.stream:
+        completion_gen = service.stream_chat(
+            messages=all_messages,
+            use_context=body.use_context,
+            context_filter=body.context_filter,
+        )
+        return StreamingResponse(
+            to_openai_sse_stream(
+                completion_gen.response,
+                completion_gen.sources if body.include_sources else None,
+            ),
+            media_type="text/event-stream",
+        )
+    else:
+        completion = service.chat(
+            messages=all_messages,
+            use_context=body.use_context,
+            context_filter=body.context_filter,
+        )
+        return to_openai_response(
+            completion.response, completion.sources if body.include_sources else None
+        )

private_gpt/server/chat/chat_service.py

@@ -0,0 +1,210 @@
+from dataclasses import dataclass
+
+from injector import inject, singleton
+from llama_index.core.chat_engine import ContextChatEngine, SimpleChatEngine
+from llama_index.core.chat_engine.types import (
+    BaseChatEngine,
+)
+from llama_index.core.indices import VectorStoreIndex
+from llama_index.core.indices.postprocessor import MetadataReplacementPostProcessor
+from llama_index.core.llms import ChatMessage, MessageRole
+from llama_index.core.postprocessor import (
+    SentenceTransformerRerank,
+    SimilarityPostprocessor,
+)
+from llama_index.core.storage import StorageContext
+from llama_index.core.types import TokenGen
+from pydantic import BaseModel
+
+from private_gpt.components.embedding.embedding_component import EmbeddingComponent
+from private_gpt.components.llm.llm_component import LLMComponent
+from private_gpt.components.node_store.node_store_component import NodeStoreComponent
+from private_gpt.components.vector_store.vector_store_component import (
+    VectorStoreComponent,
+)
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.server.chunks.chunks_service import Chunk
+from private_gpt.settings.settings import Settings
+
+
+class Completion(BaseModel):
+    response: str
+    sources: list[Chunk] | None = None
+
+
+class CompletionGen(BaseModel):
+    response: TokenGen
+    sources: list[Chunk] | None = None
+
+
+@dataclass
+class ChatEngineInput:
+    system_message: ChatMessage | None = None
+    last_message: ChatMessage | None = None
+    chat_history: list[ChatMessage] | None = None
+
+    @classmethod
+    def from_messages(cls, messages: list[ChatMessage]) -> "ChatEngineInput":
+        # Detect if there is a system message, extract the last message and chat history
+        system_message = (
+            messages[0]
+            if len(messages) > 0 and messages[0].role == MessageRole.SYSTEM
+            else None
+        )
+        last_message = (
+            messages[-1]
+            if len(messages) > 0 and messages[-1].role == MessageRole.USER
+            else None
+        )
+        # Remove from messages list the system message and last message,
+        # if they exist. The rest is the chat history.
+        if system_message:
+            messages.pop(0)
+        if last_message:
+            messages.pop(-1)
+        chat_history = messages if len(messages) > 0 else None
+
+        return cls(
+            system_message=system_message,
+            last_message=last_message,
+            chat_history=chat_history,
+        )
+
+
+@singleton
+class ChatService:
+    settings: Settings
+
+    @inject
+    def __init__(
+        self,
+        settings: Settings,
+        llm_component: LLMComponent,
+        vector_store_component: VectorStoreComponent,
+        embedding_component: EmbeddingComponent,
+        node_store_component: NodeStoreComponent,
+    ) -> None:
+        self.settings = settings
+        self.llm_component = llm_component
+        self.embedding_component = embedding_component
+        self.vector_store_component = vector_store_component
+        self.storage_context = StorageContext.from_defaults(
+            vector_store=vector_store_component.vector_store,
+            docstore=node_store_component.doc_store,
+            index_store=node_store_component.index_store,
+        )
+        self.index = VectorStoreIndex.from_vector_store(
+            vector_store_component.vector_store,
+            storage_context=self.storage_context,
+            llm=llm_component.llm,
+            embed_model=embedding_component.embedding_model,
+            show_progress=True,
+        )
+
+    def _chat_engine(
+        self,
+        system_prompt: str | None = None,
+        use_context: bool = False,
+        context_filter: ContextFilter | None = None,
+    ) -> BaseChatEngine:
+        settings = self.settings
+        if use_context:
+            vector_index_retriever = self.vector_store_component.get_retriever(
+                index=self.index,
+                context_filter=context_filter,
+                similarity_top_k=self.settings.rag.similarity_top_k,
+            )
+            node_postprocessors = [
+                MetadataReplacementPostProcessor(target_metadata_key="window"),
+                SimilarityPostprocessor(
+                    similarity_cutoff=settings.rag.similarity_value
+                ),
+            ]
+
+            if settings.rag.rerank.enabled:
+                rerank_postprocessor = SentenceTransformerRerank(
+                    model=settings.rag.rerank.model, top_n=settings.rag.rerank.top_n
+                )
+                node_postprocessors.append(rerank_postprocessor)
+
+            return ContextChatEngine.from_defaults(
+                system_prompt=system_prompt,
+                retriever=vector_index_retriever,
+                llm=self.llm_component.llm,  # Takes no effect at the moment
+                node_postprocessors=node_postprocessors,
+            )
+        else:
+            return SimpleChatEngine.from_defaults(
+                system_prompt=system_prompt,
+                llm=self.llm_component.llm,
+            )
+
+    def stream_chat(
+        self,
+        messages: list[ChatMessage],
+        use_context: bool = False,
+        context_filter: ContextFilter | None = None,
+    ) -> CompletionGen:
+        chat_engine_input = ChatEngineInput.from_messages(messages)
+        last_message = (
+            chat_engine_input.last_message.content
+            if chat_engine_input.last_message
+            else None
+        )
+        system_prompt = (
+            chat_engine_input.system_message.content
+            if chat_engine_input.system_message
+            else None
+        )
+        chat_history = (
+            chat_engine_input.chat_history if chat_engine_input.chat_history else None
+        )
+
+        chat_engine = self._chat_engine(
+            system_prompt=system_prompt,
+            use_context=use_context,
+            context_filter=context_filter,
+        )
+        streaming_response = chat_engine.stream_chat(
+            message=last_message if last_message is not None else "",
+            chat_history=chat_history,
+        )
+        sources = [Chunk.from_node(node) for node in streaming_response.source_nodes]
+        completion_gen = CompletionGen(
+            response=streaming_response.response_gen, sources=sources
+        )
+        return completion_gen
+
+    def chat(
+        self,
+        messages: list[ChatMessage],
+        use_context: bool = False,
+        context_filter: ContextFilter | None = None,
+    ) -> Completion:
+        chat_engine_input = ChatEngineInput.from_messages(messages)
+        last_message = (
+            chat_engine_input.last_message.content
+            if chat_engine_input.last_message
+            else None
+        )
+        system_prompt = (
+            chat_engine_input.system_message.content
+            if chat_engine_input.system_message
+            else None
+        )
+        chat_history = (
+            chat_engine_input.chat_history if chat_engine_input.chat_history else None
+        )
+
+        chat_engine = self._chat_engine(
+            system_prompt=system_prompt,
+            use_context=use_context,
+            context_filter=context_filter,
+        )
+        wrapped_response = chat_engine.chat(
+            message=last_message if last_message is not None else "",
+            chat_history=chat_history,
+        )
+        sources = [Chunk.from_node(node) for node in wrapped_response.source_nodes]
+        completion = Completion(response=wrapped_response.response, sources=sources)
+        return completion

private_gpt/server/chunks/chunks_router.py

@@ -0,0 +1,55 @@
+from typing import Literal
+
+from fastapi import APIRouter, Depends, Request
+from pydantic import BaseModel, Field
+
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.server.chunks.chunks_service import Chunk, ChunksService
+from private_gpt.server.utils.auth import authenticated
+
+chunks_router = APIRouter(prefix="/v1", dependencies=[Depends(authenticated)])
+
+
+class ChunksBody(BaseModel):
+    text: str = Field(examples=["Q3 2023 sales"])
+    context_filter: ContextFilter | None = None
+    limit: int = 10
+    prev_next_chunks: int = Field(default=0, examples=[2])
+
+
+class ChunksResponse(BaseModel):
+    object: Literal["list"]
+    model: Literal["private-gpt"]
+    data: list[Chunk]
+
+
+@chunks_router.post("/chunks", tags=["Context Chunks"])
+def chunks_retrieval(request: Request, body: ChunksBody) -> ChunksResponse:
+    """Given a `text`, returns the most relevant chunks from the ingested documents.
+
+    The returned information can be used to generate prompts that can be
+    passed to `/completions` or `/chat/completions` APIs. Note: it is usually a very
+    fast API, because only the Embeddings model is involved, not the LLM. The
+    returned information contains the relevant chunk `text` together with the source
+    `document` it is coming from. It also contains a score that can be used to
+    compare different results.
+
+    The max number of chunks to be returned is set using the `limit` param.
+
+    Previous and next chunks (pieces of text that appear right before or after in the
+    document) can be fetched by using the `prev_next_chunks` field.
+
+    The documents being used can be filtered using the `context_filter` and passing
+    the document IDs to be used. Ingested documents IDs can be found using
+    `/ingest/list` endpoint. If you want all ingested documents to be used,
+    remove `context_filter` altogether.
+    """
+    service = request.state.injector.get(ChunksService)
+    results = service.retrieve_relevant(
+        body.text, body.context_filter, body.limit, body.prev_next_chunks
+    )
+    return ChunksResponse(
+        object="list",
+        model="private-gpt",
+        data=results,
+    )

private_gpt/server/chunks/chunks_service.py

@@ -0,0 +1,125 @@
+from typing import TYPE_CHECKING, Literal
+
+from injector import inject, singleton
+from llama_index.core.indices import VectorStoreIndex
+from llama_index.core.schema import NodeWithScore
+from llama_index.core.storage import StorageContext
+from pydantic import BaseModel, Field
+
+from private_gpt.components.embedding.embedding_component import EmbeddingComponent
+from private_gpt.components.llm.llm_component import LLMComponent
+from private_gpt.components.node_store.node_store_component import NodeStoreComponent
+from private_gpt.components.vector_store.vector_store_component import (
+    VectorStoreComponent,
+)
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.server.ingest.model import IngestedDoc
+
+if TYPE_CHECKING:
+    from llama_index.core.schema import RelatedNodeInfo
+
+
+class Chunk(BaseModel):
+    object: Literal["context.chunk"]
+    score: float = Field(examples=[0.023])
+    document: IngestedDoc
+    text: str = Field(examples=["Outbound sales increased 20%, driven by new leads."])
+    previous_texts: list[str] | None = Field(
+        default=None,
+        examples=[["SALES REPORT 2023", "Inbound didn't show major changes."]],
+    )
+    next_texts: list[str] | None = Field(
+        default=None,
+        examples=[
+            [
+                "New leads came from Google Ads campaign.",
+                "The campaign was run by the Marketing Department",
+            ]
+        ],
+    )
+
+    @classmethod
+    def from_node(cls: type["Chunk"], node: NodeWithScore) -> "Chunk":
+        doc_id = node.node.ref_doc_id if node.node.ref_doc_id is not None else "-"
+        return cls(
+            object="context.chunk",
+            score=node.score or 0.0,
+            document=IngestedDoc(
+                object="ingest.document",
+                doc_id=doc_id,
+                doc_metadata=node.metadata,
+            ),
+            text=node.get_content(),
+        )
+
+
+@singleton
+class ChunksService:
+    @inject
+    def __init__(
+        self,
+        llm_component: LLMComponent,
+        vector_store_component: VectorStoreComponent,
+        embedding_component: EmbeddingComponent,
+        node_store_component: NodeStoreComponent,
+    ) -> None:
+        self.vector_store_component = vector_store_component
+        self.llm_component = llm_component
+        self.embedding_component = embedding_component
+        self.storage_context = StorageContext.from_defaults(
+            vector_store=vector_store_component.vector_store,
+            docstore=node_store_component.doc_store,
+            index_store=node_store_component.index_store,
+        )
+
+    def _get_sibling_nodes_text(
+        self, node_with_score: NodeWithScore, related_number: int, forward: bool = True
+    ) -> list[str]:
+        explored_nodes_texts = []
+        current_node = node_with_score.node
+        for _ in range(related_number):
+            explored_node_info: RelatedNodeInfo | None = (
+                current_node.next_node if forward else current_node.prev_node
+            )
+            if explored_node_info is None:
+                break
+
+            explored_node = self.storage_context.docstore.get_node(
+                explored_node_info.node_id
+            )
+
+            explored_nodes_texts.append(explored_node.get_content())
+            current_node = explored_node
+
+        return explored_nodes_texts
+
+    def retrieve_relevant(
+        self,
+        text: str,
+        context_filter: ContextFilter | None = None,
+        limit: int = 10,
+        prev_next_chunks: int = 0,
+    ) -> list[Chunk]:
+        index = VectorStoreIndex.from_vector_store(
+            self.vector_store_component.vector_store,
+            storage_context=self.storage_context,
+            llm=self.llm_component.llm,
+            embed_model=self.embedding_component.embedding_model,
+            show_progress=True,
+        )
+        vector_index_retriever = self.vector_store_component.get_retriever(
+            index=index, context_filter=context_filter, similarity_top_k=limit
+        )
+        nodes = vector_index_retriever.retrieve(text)
+        nodes.sort(key=lambda n: n.score or 0.0, reverse=True)
+
+        retrieved_nodes = []
+        for node in nodes:
+            chunk = Chunk.from_node(node)
+            chunk.previous_texts = self._get_sibling_nodes_text(
+                node, prev_next_chunks, False
+            )
+            chunk.next_texts = self._get_sibling_nodes_text(node, prev_next_chunks)
+            retrieved_nodes.append(chunk)
+
+        return retrieved_nodes

private_gpt/server/completions/__init__.py

@@ -0,0 +1 @@
+"""Deprecated Openai compatibility endpoint."""

private_gpt/server/completions/completions_router.py

@@ -0,0 +1,92 @@
+from fastapi import APIRouter, Depends, Request
+from pydantic import BaseModel
+from starlette.responses import StreamingResponse
+
+from private_gpt.open_ai.extensions.context_filter import ContextFilter
+from private_gpt.open_ai.openai_models import (
+    OpenAICompletion,
+    OpenAIMessage,
+)
+from private_gpt.server.chat.chat_router import ChatBody, chat_completion
+from private_gpt.server.utils.auth import authenticated
+
+completions_router = APIRouter(prefix="/v1", dependencies=[Depends(authenticated)])
+
+
+class CompletionsBody(BaseModel):
+    prompt: str
+    system_prompt: str | None = None
+    use_context: bool = False
+    context_filter: ContextFilter | None = None
+    include_sources: bool = True
+    stream: bool = False
+
+    model_config = {
+        "json_schema_extra": {
+            "examples": [
+                {
+                    "prompt": "How do you fry an egg?",
+                    "system_prompt": "You are a rapper. Always answer with a rap.",
+                    "stream": False,
+                    "use_context": False,
+                    "include_sources": False,
+                }
+            ]
+        }
+    }
+
+
+@completions_router.post(
+    "/completions",
+    response_model=None,
+    summary="Completion",
+    responses={200: {"model": OpenAICompletion}},
+    tags=["Contextual Completions"],
+    openapi_extra={
+        "x-fern-streaming": {
+            "stream-condition": "stream",
+            "response": {"$ref": "#/components/schemas/OpenAICompletion"},
+            "response-stream": {"$ref": "#/components/schemas/OpenAICompletion"},
+        }
+    },
+)
+def prompt_completion(
+    request: Request, body: CompletionsBody
+) -> OpenAICompletion | StreamingResponse:
+    """We recommend most users use our Chat completions API.
+
+    Given a prompt, the model will return one predicted completion.
+
+    Optionally include a `system_prompt` to influence the way the LLM answers.
+
+    If `use_context`
+    is set to `true`, the model will use context coming from the ingested documents
+    to create the response. The documents being used can be filtered using the
+    `context_filter` and passing the document IDs to be used. Ingested documents IDs
+    can be found using `/ingest/list` endpoint. If you want all ingested documents to
+    be used, remove `context_filter` altogether.
+
+    When using `'include_sources': true`, the API will return the source Chunks used
+    to create the response, which come from the context provided.
+
+    When using `'stream': true`, the API will return data chunks following [OpenAI's
+    streaming model](https://platform.openai.com/docs/api-reference/chat/streaming):
+    ```
+    {"id":"12345","object":"completion.chunk","created":1694268190,
+    "model":"private-gpt","choices":[{"index":0,"delta":{"content":"Hello"},
+    "finish_reason":null}]}
+    ```
+    """
+    messages = [OpenAIMessage(content=body.prompt, role="user")]
+    # If system prompt is passed, create a fake message with the system prompt.
+    if body.system_prompt:
+        messages.insert(0, OpenAIMessage(content=body.system_prompt, role="system"))
+
+    chat_body = ChatBody(
+        messages=messages,
+        use_context=body.use_context,
+        stream=body.stream,
+        include_sources=body.include_sources,
+        context_filter=body.context_filter,
+    )
+    return chat_completion(request, chat_body)