```
├── .github/
├── ISSUE_TEMPLATE/
├── bug_report.yml (400 tokens)
├── config.yml (100 tokens)
├── assets/
├── banner.png
├── streamlit_chat.png
├── video-th.png
├── pull_request_template.md (200 tokens)
├── workflows/
├── python-lint.yml (200 tokens)
├── .gitignore (600 tokens)
├── .vscode/
├── settings.json (100 tokens)
├── CODE_OF_CONDUCT.md (1200 tokens)
├── CONTRIBUTING.md (400 tokens)
├── LICENSE (omitted)
├── README.md (1800 tokens)
├── assets/
├── aws-fargate-logo.svg (700 tokens)
├── azure-logo.svg (400 tokens)
├── gcp-logo.svg (100 tokens)
├── render.png
├── cookbooks/
├── self-rag-agents/
├── pathway_deploy_langgraph_agents.ipynb (41.4k tokens)
├── pathway_langgraph_agentic_rag.ipynb (40.5k tokens)
├── pyproject.toml (400 tokens)
├── setup.cfg
├── templates/
├── adaptive_rag/
├── .env.example
├── Dockerfile (100 tokens)
├── README.md (1900 tokens)
├── app.py (500 tokens)
├── app.yaml (900 tokens)
├── data/
├── IdeanomicsInc_20160330_10-K_EX-10.26_9512211_EX-10.26_Content License Agreement.pdf
├── requirements.txt
├── document_indexing/
├── Dockerfile (100 tokens)
├── README.md (2.4k tokens)
├── app.py (500 tokens)
├── app.yaml (700 tokens)
├── docker-compose.yml (100 tokens)
├── files-for-indexing/
├── 2023q2-alphabet-earnings-release.pdf
├── Always up-to-date Vector Data Indexing pipeline _ Pathway.pdf
├── Build an LLM App _ Pathway.pdf
├── Launching Pathway + LlamaIndex.pdf
├── Realtime Classification with Nearest Neighbors (1_2) _ Pathway.pdf
├── Realtime Twitter Analysis App _ Pathway.pdf
├── Use LLMs to Ingest Raw Text into DB _ Pathway.pdf
├── arxiv 2307.13116.pdf
├── pw.io.http _ Pathway.pdf
├── requirements.txt
├── document_store_mcp_server/
├── .env.example
├── Dockerfile (100 tokens)
├── README.md (2.4k tokens)
├── __init__.py
├── app.py (300 tokens)
├── app.yaml (700 tokens)
├── docker-compose.yml
├── files-for-indexing/
├── 2023q2-alphabet-earnings-release.pdf
├── Always up-to-date Vector Data Indexing pipeline _ Pathway.pdf
├── Build an LLM App _ Pathway.pdf
├── Launching Pathway + LlamaIndex.pdf
├── Realtime Classification with Nearest Neighbors (1_2) _ Pathway.pdf
├── Realtime Twitter Analysis App _ Pathway.pdf
├── Use LLMs to Ingest Raw Text into DB _ Pathway.pdf
├── arxiv 2307.13116.pdf
├── pw.io.http _ Pathway.pdf
├── requirements.txt
├── drive_alert/
├── Dockerfile
├── README.md (1300 tokens)
├── __init__.py
├── app.py (2k tokens)
├── docker-compose.yml (100 tokens)
├── drive_alert_demo.gif
├── ui/
├── Dockerfile
├── server.py (400 tokens)
├── multimodal_rag/
├── .env.example
├── Dockerfile (100 tokens)
├── README.md (3k tokens)
├── app.py (500 tokens)
├── app.yaml (1000 tokens)
├── data/
├── 20230203_alphabet_10K.pdf
├── gpt4o.gif
├── gpt4o_with_pathway_comparison.gif
├── requirements.txt
├── private_rag/
├── Dockerfile (100 tokens)
├── README.md (2.7k tokens)
├── app.py (500 tokens)
├── app.yaml (900 tokens)
├── data/
├── IdeanomicsInc_20160330_10-K_EX-10.26_9512211_EX-10.26_Content License Agreement.pdf
├── private_rag.png
├── requirements.txt
├── question_answering_rag/
├── .env.example
├── Dockerfile (100 tokens)
├── README.md (4.3k tokens)
├── app.py (500 tokens)
├── app.yaml (900 tokens)
├── data/
├── IdeanomicsInc_20160330_10-K_EX-10.26_9512211_EX-10.26_Content License Agreement.pdf
├── docker-compose.yml (100 tokens)
├── requirements.txt
├── ui/
├── .streamlit/
├── config.toml
├── Dockerfile (100 tokens)
├── favicon.ico
├── requirements.txt
├── static/
├── pathway-logo-black.png
├── ui.py (1100 tokens)
├── slides_ai_search/
├── .dockerignore
├── .env.example
├── .gitignore
├── Dockerfile (100 tokens)
├── README.md (3.5k tokens)
├── ai-slides-diagram.svg (855.1k tokens)
├── app.py (700 tokens)
├── app.yaml (900 tokens)
├── docker-compose.yml (200 tokens)
├── nginx/
├── Dockerfile
├── nginx.conf (100 tokens)
├── pathway_slides_ai_search/
├── __init__.py (800 tokens)
├── requirements.txt
├── ui/
├── .streamlit/
├── config.toml
├── Dockerfile (100 tokens)
├── favicon.ico
├── requirements.txt
├── static/
├── Google_Drive_logo.png
├── sharepoint.png
├── ui.py (2.4k tokens)
├── unstructured_to_sql_on_the_fly/
├── Dockerfile (100 tokens)
├── README.md (1000 tokens)
├── Unstructured_to_SQL_diagram.png
├── __init__.py
├── app.py (2.3k tokens)
├── data/
├── quarterly_earnings/
├── 2023q2-alphabet-earnings-release.pdf
├── FY22_Q4_Consolidated_Financial_Statements.pdf
├── FY23_Q1_Consolidated_Financial_Statements.pdf
├── FY23_Q2_Consolidated_Financial_Statements.pdf
├── FY23_Q3_Consolidated_Financial_Statements.pdf
├── Meta-03-31-2023-Exhibit-99-1-FINAL-v2.pdf
├── goog-exhibit-99-1-q1-2023-19.pdf
├── docker-compose.yml (200 tokens)
├── postgres/
├── init-db.sql (100 tokens)
├── requirements.txt
├── ui/
├── Dockerfile
├── server.py (500 tokens)
├── unstructured_to_sql_demo.gif
```
## /.github/ISSUE_TEMPLATE/bug_report.yml
```yml path="/.github/ISSUE_TEMPLATE/bug_report.yml"
name: Bug Report
description: File a bug report
title: "[Bug]: "
labels: ["bug"]
body:
- type: markdown
attributes:
value: |
Thanks for taking the time to fill out this bug report!
- type: textarea
id: what-happened
attributes:
label: Steps to reproduce
description: What happened and how to reproduce it?
placeholder: Tell us what you see!
validations:
required: true
- type: textarea
id: logs
attributes:
label: Relevant log output
description: Please copy and paste any relevant log output. This will be automatically formatted into code, so no need for backticks.
placeholder: Tell us what you see! It would be nice if you could leave us the logs.
render: text
validations:
required: true
- type: textarea
id: what-did-you-expect
attributes:
label: What did you expect to happen?
description: Also tell us, what did you expect to happen?
placeholder: Describe your expectations.
validations:
required: true
- type: input
id: pathway_version
attributes:
label: Version
description: On which version of Pathway are you working?
validations:
required: true
- type: input
id: docker_version
attributes:
label: Docker Versions (if used)
description: On which Docker version are you working?
validations:
required: false
- type: dropdown
id: os_version
attributes:
label: OS
description: On which OS did you run Pathway?
options:
- Linux
- MacOS
validations:
required: true
- type: dropdown
id: cpu_architecture
attributes:
label: On which CPU architecture did you run Pathway?
multiple: false
options:
- x86-64
- ARM64 (AArch64, Apple silicon)
```
## /.github/ISSUE_TEMPLATE/config.yml
```yml path="/.github/ISSUE_TEMPLATE/config.yml"
blank_issues_enabled: false
contact_links:
- name: New feature? Tell us about it on GitHub Discussions
url: https://github.com/pathwaycom/pathway/discussions
about: We are using Discussions as a place to connect with other members of the Pathway community.
- name: Discord
url: https://discord.com/invite/pathway
about: Ask questions you are wondering about.
```
## /.github/assets/banner.png
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/.github/assets/banner.png
## /.github/assets/streamlit_chat.png
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/.github/assets/streamlit_chat.png
## /.github/assets/video-th.png
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/.github/assets/video-th.png
## /.github/pull_request_template.md
### Introduction
To contribute code to the Pathway project, start by discussing your proposed changes on Discord or by filing an issue.
Once approved, follow the fork + pull request model against the main branch, ensuring you've signed the contributor license agreement.
### Context
<!--- Why is this change required? What problem does it solve? -->
### How has this been tested?
<!--- Please describe in detail how you tested your changes. -->
### Types of changes
<!--- What types of changes does your code introduce? Put an `x` in all the boxes that apply: -->
- [ ] Bug fix (non-breaking change which fixes an issue)
- [ ] New feature or improvement (non-breaking change which adds functionality)
- [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
### Related issue(s):
1.
2.
3.
### Checklist:
<!--- Go over all the following points, and put an `x` in all the boxes that apply. -->
<!--- If you're unsure about any of these, don't hesitate to ask. We're here to help! -->
- [ ] My code follows the code style of this project,
- [ ] My change requires a change to the documentation,
- [ ] I described the modification in the CHANGELOG.md file.
## /.github/workflows/python-lint.yml
```yml path="/.github/workflows/python-lint.yml"
name: lint PR
on:
pull_request:
push:
branches:
- main
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main'}}
jobs:
linter:
runs-on: ubuntu-latest
strategy:
fail-fast: false
steps:
- name: Git checkout
uses: actions/checkout@v3
- name: Set up Python
uses: actions/setup-python@v5
with:
python-version: 3.11
- name: Install poetry
uses: abatilo/actions-poetry@v3
- name: Install linters
run: poetry install --no-root --only linters
- name: Run black
run: poetry run black --check .
- name: Run Flake8
run: poetry run flake8 .
- name: Run mypy
run: poetry run mypy .
- name: Run isort
run: poetry run isort --check .
```
## /.gitignore
```gitignore path="/.gitignore"
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$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
# pyenv
# For a library or package, you might want to ignore these files since the code is
# intended to run in multiple environments; otherwise, check them in:
# .python-version
# pipenv
# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
# However, in case of collaboration, if having platform-specific dependencies or dependencies
# having no cross-platform support, pipenv may install dependencies that don't work, or not
# install all needed dependencies.
#Pipfile.lock
# poetry
# Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
# This is especially recommended for binary packages to ensure reproducibility, and is more
# commonly ignored for libraries.
# https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
#poetry.lock
# pdm
# Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
#pdm.lock
# pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
# in version control.
# https://pdm.fming.dev/#use-with-ide
.pdm.toml
# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
__pypackages__/
# Celery stuff
celerybeat-schedule
celerybeat.pid
# SageMath parsed files
*.sage.py
# Environments
**/.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/
# Spyder project settings
.spyderproject
.spyproject
# Rope project settings
.ropeproject
# mkdocs documentation
/site
# mypy
.mypy_cache/
.dmypy.json
dmypy.json
# Pyre type checker
.pyre/
# pytype static type analyzer
.pytype/
# Cython debug symbols
cython_debug/
# PyCharm
# JetBrains specific template is maintained in a separate JetBrains.gitignore that can
# be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
# and can be added to the global gitignore or merged into this file. For a more nuclear
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
#.idea/
pw-env/
# llm debug
examples/ui/data/*
```
## /.vscode/settings.json
```json path="/.vscode/settings.json"
{
"python.defaultInterpreterPath": "${env:HOME}/pw-env/bin/python",
"python.formatting.provider": "none",
"editor.formatOnSave": true,
"python.linting.enabled": true,
"python.linting.flake8Enabled": true,
"python.linting.mypyEnabled": true,
"[python]": {
"files.trimTrailingWhitespace": true,
"editor.rulers": [88],
"editor.codeActionsOnSave": {
"source.organizeImports": "explicit"
},
"editor.defaultFormatter": "ms-python.black-formatter"
},
"git.autofetchPeriod": 315360000
}
```
## /CODE_OF_CONDUCT.md
# Pathway Code of Conduct
In the interest of fostering an open and welcoming environment, we as
contributors and maintainers pledge to make participation in our project and our
community a harassment-free experience for everyone, regardless of age, body
size, disability, health condition, ethnicity, gender identity and expression, level of
experience, nationality, country of origin, personal appearance, race, religion, or sexual identity
and orientation, and any other criteria falling under discriminatory practices under the Law of France.
## Our Standards
Examples of behavior that contributes to creating a positive environment include:
* Using welcoming and inclusive language.
* Being respectful of differing viewpoints and experiences.
* Gracefully accepting constructive criticism.
* Focusing on what is best for the community.
* Showing empathy towards other community members.
Examples of unacceptable behavior by participants include:
* The use of sexualized language or imagery and unwelcome sexual attention or
advances.
* Trolling, insulting/derogatory comments, and personal or political attacks.
* Public or private harassment.
* Publishing others' private information, such as a physical or electronic
address, without explicit permission.
* Conduct which could reasonably be considered inappropriate for the forum in
which it occurs.
All Pathway forums and spaces are meant for professional interactions, and any behavior which could reasonably be considered inappropriate in a professional setting is unacceptable.
## Our Responsibilities
Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
## Scope
This Code of Conduct applies to all content on pathway.com, Pathway’s GitHub organization, or any other official Pathway web presence allowing for community interactions, as well as at all official Pathway events, whether offline or online.
The Code of Conduct also applies within project spaces and in public spaces whenever an individual is representing Pathway or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed or de facto representative at an online or offline event.
## Conflict Resolution
Conflicts in an open source project can take many forms, from someone having a bad day and using harsh and hurtful language in the issue queue, to more serious instances such as sexist/racist statements or threats of violence, and everything in between.
If the behavior is threatening or harassing, or for other reasons requires immediate escalation, please see below.
However, for the vast majority of issues, we aim to empower individuals to first resolve conflicts themselves, asking for help when needed, and only after that fails to escalate further. This approach gives people more control over the outcome of their dispute.
If you are experiencing or witnessing conflict, we ask you to use the following escalation strategy to address the conflict:
1. If you so wish and if you do not feel threatened or at risk of any form of personal abuse, address the perceived conflict directly with those involved, preferably in a real-time medium.
2. If this fails, get a third party (e.g. a mutual friend, and/or someone with
background on the issue, but not involved in the conflict) to intercede.
3. If you are still unable to resolve the conflict, and you believe it rises to
harassment or another code of conduct violation, report it.
Please note that if you are experiencing or witnessing a discriminatory practice that would be susceptible to be condemned by law, we ask you to directly escalate to 3.
## Reporting Violations
Violations of the Code of Conduct can be reported to Pathway via email to code_of_conduct@pathway.com. Project maintainers will determine whether the Code of Conduct was violated, and will issue an appropriate sanction, possibly including a written warning or expulsion from the project, project sponsored spaces, or project forums. We ask that you make a good-faith effort to resolve your conflict via the conflict resolution policy before submitting a report.
Violations of the Code of Conduct can occur in any setting, even those unrelated to the project. We will only consider complaints about conduct that has occurred within one year of the report.
## Enforcement
If the Project maintainers receive a report alleging a violation of the Code of Conduct, the Project maintainers will notify the accused of the report, and provide them an opportunity to discuss the report before a sanction is issued. The Project maintainers will do their utmost to keep the reporter anonymous. If the act is ongoing (such as someone engaging in harassment), or involves a threat to anyone's safety (e.g. threats of violence), the Project maintainers may issue sanctions without notice.
## Attribution
This Code of Conduct is adapted from the Tensorflow Code of Conduct, and based on Contributor Covenant, version 1.4, available at https://contributor-covenant.org/version/1/4, and includes some aspects of the Geek Feminism Code of Conduct and the Drupal Code of Conduct.
## /CONTRIBUTING.md
# Contributing to Pathway
Welcome! This guide is intended to help developers that are new to the community
to make contributions to the Pathway project. Please be sure to also read our [code of conduct](CODE_OF_CONDUCT.md). We work hard to make this a welcoming community for all, and we're excited to have you on board!
## The basics:
* Use the [Issue Tracker](https://github.com/pathwaycom/llm-app/issues) to
report bugs, crashes, performance issues, etc... Please include as much detail
as possible.
* [Discussions](https://github.com/pathwaycom/pathway/discussions) is a great
venue to ask questions, start a design discussion, or post an RFC.
* [Discord](https://discord.com/invite/pathway) is our main gathering place - jump in and ask us anything!
## Contributing Code
We are eager to build Pathway together with the community and are excited to have you here!
Important: Before you contribute any changes to our repositories (make a Pull Request) please get in touch with us and explain what you want to work on. You can do so by filing an issue, or asking us on Discord. That way you can get coding advice and speedier reviews. If you start to work on an issue before hearing back from a Pathway core team engineer there is a chance that we will not be able to accept your changes - something we all want to avoid!
We use a standard GitHub fork + pull request model for merging and reviewing
changes. New pull requests should be made against the main upstream branch.
After a pull request is opened it will be reviewed, and merged after
passing continuous integration tests and being accepted by a project or
sub-system maintainer.
We ask that developers sign our [contributor license
agreement](https://cla-assistant.io/pathwaycom/llm-app). The
process of signing the CLA is automated, and you'll be prompted with instructions
the first time you submit a pull request to the project.
## /README.md
<div align="center">
# Pathway AI Pipelines
<a href="https://trendshift.io/repositories/4400" target="_blank"><img src="https://trendshift.io/api/badge/repositories/4400" alt="pathwaycom%2Fllm-app | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
[](https://discord.gg/pathway)
[](https://x.com/intent/follow?screen_name=pathway_com)
</div>
Pathway's **AI Pipelines** allow you to quickly put in production AI applications that offer **high-accuracy RAG and AI enterprise search at scale** using the most **up-to-date knowledge** available in your data sources. It provides you ready-to-deploy **LLM (Large Language Model) App Templates**. You can test them on your own machine and deploy on-cloud (GCP, AWS, Azure, Render,...) or on-premises.
The apps connect and sync (all new data additions, deletions, updates) with data sources on your **file system, Google Drive, Sharepoint, S3, Kafka, PostgreSQL, real-time data APIs**. They come with no infrastructure dependencies that would need a separate setup. They include **built-in data indexing** enabling vector search, hybrid search, and full-text search - all done in-memory, with cache.
## Application Templates
The application templates provided in this repo scale up to **millions of pages of documents**. Some of them are optimized for simplicity, some are optimized for amazing accuracy. Pick the one that suits you best. You can use it out of the box, or change some steps of the pipeline - for example, if you would like to add a new data source, or change a Vector Index into a Hybrid Index, it's just a one-line change.
| Application (template) | Description |
| --------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [`Question-Answering RAG App`](templates/question_answering_rag/) | Basic end-to-end RAG app. A question-answering pipeline that uses the GPT model of choice to provide answers to queries to your documents (PDF, DOCX,...) on a live connected data source (files, Google Drive, Sharepoint,...). You can also try out a [demo REST endpoint](https://pathway.com/solutions/rag-pipelines#try-it-out). |
| [`Live Document Indexing (Vector Store / Retriever)`](templates/document_indexing/) | A real-time document indexing pipeline for RAG that acts as a vector store service. It performs live indexing on your documents (PDF, DOCX,...) from a connected data source (files, Google Drive, Sharepoint,...). It can be used with any frontend, or integrated as a retriever backend for a [Langchain](https://pathway.com/blog/langchain-integration) or [Llamaindex](https://pathway.com/blog/llamaindex-pathway) application. You can also try out a [demo REST endpoint](https://pathway.com/solutions/ai-contract-management#try-it-out). |
| [`Multimodal RAG pipeline with GPT4o`](templates/multimodal_rag/) | Multimodal RAG using GPT-4o in the parsing stage to index PDFs and other documents from a connected data source files, Google Drive, Sharepoint,...). It is perfect for extracting information from unstructured financial documents in your folders (including charts and tables), updating results as documents change or new ones arrive.|
| [`Unstructured-to-SQL pipeline + SQL question-answering`](templates/unstructured_to_sql_on_the_fly/) | A RAG example which connects to unstructured financial data sources (financial report PDFs), structures the data into SQL, and loads it into a PostgreSQL table. It also answers natural language user queries to these financial documents by translating them into SQL using an LLM and executing the query on the PostgreSQL table. |
| [`Adaptive RAG App`](templates/adaptive_rag/) | A RAG application using Adaptive RAG, a technique developed by Pathway to reduce token cost in RAG up to 4x while maintaining accuracy. |
| [`Private RAG App with Mistral and Ollama`](templates/private_rag/) | A fully private (local) version of the `question_answering_rag` RAG pipeline using Pathway, Mistral, and Ollama. |
| [`Slides AI Search App`](templates/slides_ai_search/) | An indexing pipeline for retrieving slides. It performs multi-modal of PowerPoint and PDF and maintains live index of your slides."|
## How do these AI Pipelines work?
The apps can be run as **Docker containers**, and expose an **HTTP API** to connect the frontend. To allow quick testing and demos, some app templates also include an optional Streamlit UI which connects to this API.
The apps rely on the [Pathway Live Data framework](https://github.com/pathwaycom/pathway) for data source synchronization and for serving API requests (Pathway is a standalone Python library with a Rust engine built into it). They bring you a **simple and unified application logic** for back-end, embedding, retrieval, LLM tech stack. There is no need to integrate and maintain separate modules for your Gen AI app: ~Vector Database (e.g. Pinecone/Weaviate/Qdrant) + Cache (e.g. Redis) + API Framework (e.g. Fast API)~. Pathway's default choice of **built-in vector index** is based on the lightning-fast [usearch](https://github.com/unum-cloud/usearch) library, and **hybrid full-text indexes** make use of [Tantivy](https://github.com/quickwit-oss/tantivy) library. Everything works out of the box.
## Getting started
Each of the [App templates](templates/) in this repo contains a README.md with instructions on how to run it.
You can also find [more ready-to-run code templates](https://pathway.com/developers/templates/) on the Pathway website.
## Some visual highlights
Effortlessly extract and organize table and chart data from PDFs, docs, and more with multimodal RAG - in real-time:
(Check out [`Multimodal RAG pipeline with GPT4o`](templates/multimodal_rag/) to see the whole pipeline in the works. You may also check out the [`Unstructured-to-SQL pipeline`](templates/unstructured_to_sql_on_the_fly/) for a minimal example that works with non-multimodal models as well.)
Automated real-time knowledge mining and alerting:
(Check out the [`Alerting when answers change on Google Drive`](https://github.com/pathwaycom/llm-app/tree/main/templates/drive_alert) app example.)
### Do-it-Yourself Videos
▶️ [An introduction to building LLM apps with Pathway](https://www.youtube.com/watch?v=kcrJSk00duw) - by [Jan Chorowski](https://scholar.google.com/citations?user=Yc94070AAAAJ)
▶️ [Let's build a real-world LLM app in 11 minutes](https://www.youtube.com/watch?v=k1XGo7ts4tI) - by [Pau Labarta Bajo](https://substack.com/@paulabartabajo)
## Troubleshooting
To provide feedback or report a bug, please [raise an issue on our issue tracker](https://github.com/pathwaycom/pathway/issues).
## Contributing
Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code cleanup, testing, or code reviews, is very much encouraged to do so. If this is your first contribution to a GitHub project, here is a [Get Started Guide](https://docs.github.com/en/get-started/quickstart/contributing-to-projects).
If you'd like to make a contribution that needs some more work, just raise your hand on the [Pathway Discord server](https://discord.com/invite/pathway) (#get-help) and let us know what you are planning!
## Supported and maintained by
<p align="center">
<a href="https://github.com/pathwaycom/"><img src="https://pathway.com/logo-light.svg" alt="Pathway"/></a>
</p>
<p align="center">
<a href="https://pathway.com/solutions/llm-app">
<img src="https://img.shields.io/badge/See%20Pathway's%20offering%20for%20AI%20applications-0000FF" alt="See Pathway's offering for AI applications"/>
</a>
</p>
## /assets/aws-fargate-logo.svg
```svg path="/assets/aws-fargate-logo.svg"
<?xml version="1.0" encoding="UTF-8"?>
<svg width="80px" height="80px" viewBox="0 0 80 80" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<!-- Generator: Sketch 64 (93537) - https://sketch.com -->
<title>Icon-Architecture/64/Arch_AWS-Fargate_64</title>
<desc>Created with Sketch.</desc>
<defs>
<linearGradient x1="0%" y1="100%" x2="100%" y2="0%" id="linearGradient-1">
<stop stop-color="#C8511B" offset="0%"></stop>
<stop stop-color="#FF9900" offset="100%"></stop>
</linearGradient>
</defs>
<g id="Icon-Architecture/64/Arch_AWS-Fargate_64" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
<g id="Icon-Architecture-BG/64/Compute" fill="url(#linearGradient-1)">
<rect id="Rectangle" x="0" y="0" width="80" height="80"></rect>
</g>
<g id="Icon-Service/64/AWS-Fargate" transform="translate(8.000000, 8.000000)" fill="#FFFFFF">
<path d="M44,48.523 L44,43.677 L47,42.477 L47,47.323 L44,48.523 Z M39,42.477 L42,43.677 L42,48.523 L39,47.323 L39,42.477 Z M33,55.523 L33,50.677 L36,49.477 L36,54.323 L33,55.523 Z M28,49.477 L31,50.677 L31,55.523 L28,54.323 L28,49.477 Z M22,48.523 L22,43.677 L25,42.477 L25,47.323 L22,48.523 Z M17,42.477 L20,43.677 L20,48.523 L17,47.323 L17,42.477 Z M21,40.077 L23.308,41 L21,41.923 L18.692,41 L21,40.077 Z M32,47.077 L34.308,48 L32,48.923 L29.692,48 L32,47.077 Z M43,40.077 L45.308,41 L43,41.923 L40.692,41 L43,40.077 Z M48.371,40.071 L43.371,38.071 C43.133,37.977 42.867,37.977 42.629,38.071 L37.629,40.071 C37.249,40.224 37,40.591 37,41 L37,46.923 L32.371,45.071 C32.133,44.977 31.867,44.977 31.629,45.071 L27,46.923 L27,41 C27,40.591 26.751,40.224 26.371,40.071 L21.371,38.071 C21.133,37.977 20.867,37.977 20.629,38.071 L15.629,40.071 C15.249,40.224 15,40.591 15,41 L15,48 C15,48.409 15.249,48.776 15.629,48.929 L20.629,50.929 C20.748,50.976 20.874,51 21,51 C21.126,51 21.252,50.976 21.371,50.929 L26,49.077 L26,55 C26,55.409 26.249,55.776 26.629,55.929 L31.629,57.929 C31.748,57.976 31.874,58 32,58 C32.126,58 32.252,57.976 32.371,57.929 L37.371,55.929 C37.751,55.776 38,55.409 38,55 L38,49.077 L42.629,50.929 C42.748,50.976 42.874,51 43,51 C43.126,51 43.252,50.976 43.371,50.929 L48.371,48.929 C48.751,48.776 49,48.409 49,48 L49,41 C49,40.591 48.751,40.224 48.371,40.071 L48.371,40.071 Z M57,27.938 C57,33.783 44.119,36.938 32,36.938 C19.881,36.938 7,33.783 7,27.938 C7,25.147 10.095,22.69 15.715,21.019 L16.285,22.936 C11.791,24.272 9,26.189 9,27.938 C9,31.249 18.445,34.938 32,34.938 C45.555,34.938 55,31.249 55,27.938 C55,26.189 52.209,24.272 47.715,22.936 L48.285,21.019 C53.905,22.69 57,25.147 57,27.938 L57,27.938 Z M32,9.071 L42.214,13 L32,16.929 L21.786,13 L32,9.071 Z M42.58,28.842 C40.694,29.686 37.528,30.669 33,30.801 L33,18.687 L44,14.457 L44,26.648 C44,27.599 43.444,28.46 42.58,28.842 L42.58,28.842 Z M20,26.648 L20,14.457 L31,18.687 L31,30.801 C26.472,30.669 23.306,29.686 21.416,28.84 C20.556,28.46 20,27.599 20,26.648 L20,26.648 Z M20.604,30.667 C22.805,31.652 26.569,32.827 32,32.827 C37.431,32.827 41.195,31.652 43.393,30.669 C44.977,29.968 46,28.39 46,26.648 L46,13 C46,12.586 45.745,12.215 45.359,12.066 L32.359,7.066 C32.127,6.978 31.873,6.978 31.641,7.066 L18.641,12.066 C18.255,12.215 18,12.586 18,13 L18,26.648 C18,28.39 19.023,29.968 20.604,30.667 L20.604,30.667 Z" id="AWS-Fargate_Icon_64_Squid"></path>
</g>
</g>
</svg>
```
## /assets/azure-logo.svg
```svg path="/assets/azure-logo.svg"
<svg width="150" height="150" viewBox="0 0 96 96" xmlns="http://www.w3.org/2000/svg">
<defs>
<linearGradient id="e399c19f-b68f-429d-b176-18c2117ff73c" x1="-1032.172" x2="-1059.213" y1="145.312" y2="65.426" gradientTransform="matrix(1 0 0 -1 1075 158)" gradientUnits="userSpaceOnUse">
<stop offset="0" stop-color="#114a8b"/>
<stop offset="1" stop-color="#0669bc"/>
</linearGradient>
<linearGradient id="ac2a6fc2-ca48-4327-9a3c-d4dcc3256e15" x1="-1023.725" x2="-1029.98" y1="108.083" y2="105.968" gradientTransform="matrix(1 0 0 -1 1075 158)" gradientUnits="userSpaceOnUse">
<stop offset="0" stop-opacity=".3"/>
<stop offset=".071" stop-opacity=".2"/>
<stop offset=".321" stop-opacity=".1"/>
<stop offset=".623" stop-opacity=".05"/>
<stop offset="1" stop-opacity="0"/>
</linearGradient>
<linearGradient id="a7fee970-a784-4bb1-af8d-63d18e5f7db9" x1="-1027.165" x2="-997.482" y1="147.642" y2="68.561" gradientTransform="matrix(1 0 0 -1 1075 158)" gradientUnits="userSpaceOnUse">
<stop offset="0" stop-color="#3ccbf4"/>
<stop offset="1" stop-color="#2892df"/>
</linearGradient>
</defs>
<path fill="url(#e399c19f-b68f-429d-b176-18c2117ff73c)" d="M33.338 6.544h26.038l-27.03 80.087a4.152 4.152 0 0 1-3.933 2.824H8.149a4.145 4.145 0 0 1-3.928-5.47L29.404 9.368a4.152 4.152 0 0 1 3.934-2.825z"/>
<path fill="#0078d4" d="M71.175 60.261h-41.29a1.911 1.911 0 0 0-1.305 3.309l26.532 24.764a4.171 4.171 0 0 0 2.846 1.121h23.38z"/>
<path fill="url(#ac2a6fc2-ca48-4327-9a3c-d4dcc3256e15)" d="M33.338 6.544a4.118 4.118 0 0 0-3.943 2.879L4.252 83.917a4.14 4.14 0 0 0 3.908 5.538h20.787a4.443 4.443 0 0 0 3.41-2.9l5.014-14.777 17.91 16.705a4.237 4.237 0 0 0 2.666.972H81.24L71.024 60.261l-29.781.007L59.47 6.544z"/>
<path fill="url(#a7fee970-a784-4bb1-af8d-63d18e5f7db9)" d="M66.595 9.364a4.145 4.145 0 0 0-3.928-2.82H33.648a4.146 4.146 0 0 1 3.928 2.82l25.184 74.62a4.146 4.146 0 0 1-3.928 5.472h29.02a4.146 4.146 0 0 0 3.927-5.472z"/>
</svg>
```
## /assets/gcp-logo.svg
```svg path="/assets/gcp-logo.svg"
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 97.75 78.62"><path d="M62.21 21.73l3.13.06 8.53-8.51.42-3.6a38.2 38.2 0 0 0-62.3 18.57c.88-.64 2.88-.16 2.88-.16l17-2.8s.88-1.44 1.31-1.35a21.19 21.19 0 0 1 29-2.21z" fill="#ea4335"/><path d="M85.79 28.27a38.32 38.32 0 0 0-11.54-18.6l-12 12a21.2 21.2 0 0 1 7.92 16.53v2.19a10.63 10.63 0 1 1 0 21.25h-21.3l-2.12 2.14v12.73l2.12 2.12h21.25a27.61 27.61 0 0 0 15.67-50.36z" fill="#4285f4"/><path d="M27.62 78.63h21.25v-17H27.62a10.54 10.54 0 0 1-4.38-1l-3.06.94-8.51 8.55-.74 2.88a27.47 27.47 0 0 0 16.69 5.63z" fill="#34a853"/><path d="M27.62 23.39A27.61 27.61 0 0 0 10.94 73l12.32-12.32a10.62 10.62 0 1 1 14-14l12.36-12.36a27.6 27.6 0 0 0-22-10.93z" fill="#fbbc04"/></svg>
```
## /assets/render.png
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/assets/render.png
## /cookbooks/self-rag-agents/pathway_deploy_langgraph_agents.ipynb
```ipynb path="/cookbooks/self-rag-agents/pathway_deploy_langgraph_agents.ipynb"
{
"cells": [
{
"cell_type": "markdown",
"id": "d58d3552-9168-4f0f-8a79-4dbaa4a01bbf",
"metadata": {
"tags": []
},
"source": [
"> The following cookbook is an adaption of the original [LangGraph cookbook](https://github.com/langchain-ai/langgraph/blob/e3ca7bb3e9d34b09633852f4d08d55f6dcd4364b/examples/rag/langgraph_self_rag.ipynb)\n",
"\n"
]
},
{
"cell_type": "markdown",
"id": "b3e94de1-ca3d-4956-b212-4a558b68062a",
"metadata": {},
"source": [
"# Power and Deploy LangGraph Agents with Pathway: A Complete Guide\n",
"This notebook demonstrates how to use Pathway to power and deploy LangGraph Agents.\n",
"\n",
"In this notebook, you will learn:\n",
"- How to build a Pathway RAG app that indexes documents from your data sources\n",
"- How to create a Hybrid Index that is powered by semantic search and BM25 index\n",
"- How to use the Pathway LangChain Retriever in a LangGraph Agent\n",
"- How to use Pathway to serve LangGraph Agents"
]
},
{
"cell_type": "markdown",
"id": "d27aa0d3-9b11-4678-8a83-b3865b26f30c",
"metadata": {},
"source": [
"# Setup\n",
"\n",
"Install the required packages and set your OpenAI API key. \n",
"OpenAI is used for embeddings during the indexing stage and to power the LangGraph agent."
]
},
{
"cell_type": "markdown",
"id": "9e055de9-723f-44e3-ad39-70cc5f8932bf",
"metadata": {},
"source": [
"Magic library is used for detecting file types in the `UnstructuredParser` module.\n",
"\n",
"If you are running this notebook on **MacOS**, you can install it with:\n",
"> `brew install libmagic`\n",
"\n",
"If you are running the notebook on **colab** or any **linux** environment, you can install it with:\n",
"> `apt-get install libmagic1`"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "209efe9f-b9d3-4760-8863-0a6052617ebd",
"metadata": {},
"outputs": [],
"source": [
"!pip install -U \"pathway[all]\""
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "b44f689e-0a62-4bbd-94e8-5bf31ece9cd9",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"!pip install -U langgraph langchain-community langchainhub langchain-openai"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3ff9da9f-66b8-4675-8c45-afa10e891462",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import os\n",
"\n",
"import json\n",
"from typing import Iterable, Literal, List\n",
"from pydantic import BaseModel, Field\n",
"\n",
"# needed for the OpenAI embedder and the LLM we will use below, you can change the embedding provider, see the documentation:\n",
"# https://pathway.com/developers/api-docs/pathway-xpacks-llm/embedders\n",
"os.environ[\"OPENAI_API_KEY\"] = \"sk-...\""
]
},
{
"cell_type": "markdown",
"id": "571d4836-4d08-4ee5-97fa-37132ccc8ef9",
"metadata": {},
"source": [
"Lets define a `DATA_PATH` folder that will store the text files/documents for indexing."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "54979252-13d3-4b94-b2ad-ec7b012aa320",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"# folder that we will gather our docs in\n",
"DATA_PATH = \"./data\"\n",
"\n",
"os.makedirs(DATA_PATH, exist_ok=True)"
]
},
{
"cell_type": "markdown",
"id": "55c2fb67-c562-47f7-a08e-6a673b4b612b",
"metadata": {},
"source": [
"## Load & save a webpage\n",
"\n",
"Define utility functions to read content from a webpage and save it locally into `DATA_PATH`.\n",
" 1. `load_page_content`: Reads the raw text from a webpage using using a `WebBaseLoader` from `langchain_community.document_loaders`.\n",
" 2. `ingest_webpage`: Saves the loaded content as a text file into the `DATA_PATH`. "
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "64538684-5b2b-460a-b3a5-1fe737fd8269",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import re\n",
"from urllib.parse import urlparse\n",
"from langchain_community.document_loaders import WebBaseLoader\n",
"\n",
"\n",
"def load_page_content(url: str) -> str:\n",
" \"\"\"Load web page content with Langchain utilities.\"\"\"\n",
" return WebBaseLoader(url).load()[0].page_content\n",
"\n",
"\n",
"def ingest_webpage(url: str) -> None:\n",
" \"\"\"Save a webpage to local `DATA_PATH` folder.\"\"\"\n",
" text_content = load_page_content(url)\n",
"\n",
" parsed_url = urlparse(url)\n",
" file_name = parsed_url.hostname + parsed_url.path.replace(\"/\", \"_\") + \".txt\"\n",
"\n",
" with open(os.path.join(DATA_PATH, file_name), \"w\", encoding=\"utf-8\") as f:\n",
" f.write(text_content)"
]
},
{
"cell_type": "markdown",
"id": "5bf15463-7da1-4d04-9f6d-423975e377cb",
"metadata": {
"tags": []
},
"source": [
"Download the Self-RAG paper"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "fff7b382-2e25-415b-a3ee-06df976b3dbd",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"ingest_webpage(\"https://arxiv.org/html/2310.11511\")"
]
},
{
"cell_type": "markdown",
"id": "62376b16-e5bc-463d-bf84-bec69a7c2035",
"metadata": {
"tags": []
},
"source": [
"# Build the Pathway indexing pipeline\n",
"\n",
"Set up Pathway to read and index the documents saved under the `DATA_PATH`.\n",
"\n",
"\n",
"1. [Connectors](https://pathway.com/developers/user-guide/connect/pathway-connectors): Use Pathway’s file reader to ingest all text files under the `DATA_PATH`.\n",
"2. [Parsers](https://pathway.com/developers/api-docs/pathway-xpacks-llm/parsers): Utilize the UnstructuredParser to parse the documents. This parser supports multiple file types, including PDF, DOCX, and PPTX.\n",
"3. [Text Splitters](https://pathway.com/developers/api-docs/pathway-xpacks-llm/splitters): Split the document content into chunks.\n",
"4. [Embedders](https://pathway.com/developers/api-docs/pathway-xpacks-llm/embedders): Use OpenAI API for embeddings."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "27c3d42a-30e1-47f8-bf3c-bd44d4a490f0",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"# host and port of the RAG app\n",
"pathway_host: str = \"0.0.0.0\"\n",
"pathway_port: int = 8000"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "094ec4e1-e972-441b-9c47-ff0ce3c8d2ff",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import pathway as pw\n",
"from pathway.stdlib.indexing import BruteForceKnnFactory, HybridIndexFactory\n",
"from pathway.stdlib.indexing.bm25 import TantivyBM25Factory\n",
"from pathway.udfs import DiskCache\n",
"from pathway.xpacks.llm import embedders, llms, parsers, splitters\n",
"from pathway.xpacks.llm.document_store import DocumentStore\n",
"from pathway.xpacks.llm.question_answering import BaseRAGQuestionAnswerer, RAGClient\n",
"\n",
"\n",
"# read the text files under the data folder, we can also read from Google Drive, Sharepoint, etc.\n",
"# See connectors documentation: https://pathway.com/developers/user-guide/connect/pathway-connectors to learn more\n",
"folder = pw.io.fs.read(\n",
" path=f\"{DATA_PATH}/*.txt\",\n",
" format=\"binary\",\n",
" with_metadata=True,\n",
")\n",
"\n",
"# list of data sources to be indexed\n",
"sources = [folder]\n",
"\n",
"# define the document processing steps\n",
"parser = parsers.UnstructuredParser()\n",
"\n",
"text_splitter = splitters.TokenCountSplitter(min_tokens=150, max_tokens=450)\n",
"\n",
"embedder = embedders.OpenAIEmbedder(cache_strategy=DiskCache())"
]
},
{
"cell_type": "markdown",
"id": "d3426ef8-90ee-4023-8094-5ce007d64b0e",
"metadata": {},
"source": [
"Hybrid index combines semantic search and keyword based BM25 search.\n",
"\n",
"[`HybridIndexFactory`](https://pathway.com/developers/api-docs/indexing#pathway.stdlib.indexing.HybridIndexFactory) combines different indexes to build an hybrid index:\n",
" 1. [BM25](https://pathway.com/developers/api-docs/indexing#pathway.stdlib.indexing.TantivyBM25) (via TantivyBM25Factory) → Keyword based BM25 search.\n",
" 2. [BruteForceKnn](https://pathway.com/developers/api-docs/indexing#pathway.stdlib.indexing.BruteForceKnn) → Vector-based semantic search"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a3f7c59b-925b-4ef3-88a8-89749479925e",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"index = HybridIndexFactory(\n",
" [\n",
" TantivyBM25Factory(),\n",
" BruteForceKnnFactory(embedder=embedder),\n",
" ]\n",
")"
]
},
{
"cell_type": "markdown",
"id": "8b55a446-cf97-4d47-b1ca-374bcf0e8a7b",
"metadata": {},
"source": [
"[DocumentStore](https://pathway.com/developers/api-docs/pathway-xpacks-llm/document_store#pathway.xpacks.llm.document_store.DocumentStore) manages document ingestion, parsing, splitting, and indexing.\n",
"\n",
"[BaseRAGQuestionAnswerer](https://pathway.com/developers/api-docs/pathway-xpacks-llm/question_answering#pathway.xpacks.llm.question_answering.BaseRAGQuestionAnswerer) creates a Pathway “RAG” application that:\n",
" * Indexes the documents (via `document_store`)\n",
" * Exposes a standard question-answering interface\n",
" * Enables us to create endpoint for the agent with Pathway."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7db0ab59-a641-43cc-901b-cc2713f03a99",
"metadata": {},
"outputs": [],
"source": [
"llm = llms.OpenAIChat(model=\"gpt-4o-mini\", cache_strategy=DiskCache())\n",
"\n",
"document_store = DocumentStore(\n",
" docs=sources, parser=parser, splitter=text_splitter, retriever_factory=index\n",
")\n",
"\n",
"# create the RAG app that will power the index, and serve the agent endpoint\n",
"rag_app = BaseRAGQuestionAnswerer(\n",
" llm=llm,\n",
" indexer=document_store,\n",
")"
]
},
{
"cell_type": "markdown",
"id": "a027e277-fba8-4198-863b-2168ae6f8505",
"metadata": {},
"source": [
"# Create the LangGraph agent"
]
},
{
"cell_type": "markdown",
"id": "cf15141f-9612-4d0d-a21b-11bd8530686a",
"metadata": {},
"source": [
"### Create the Langchain Retriever\n",
"\n",
"You can now query Pathway and access up-to-date documents for your RAG applications from LangChain using [PathwayVectorClient](https://api.python.langchain.com/en/latest/vectorstores/langchain_community.vectorstores.pathway.PathwayVectorClient.html)\n",
"\n",
"The Langchain Retriever is a wrapper around the Pathway client, you will use it in `retrieve` part stage of the Agent"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "fefffab6-d5a3-4925-a52d-5a5c574421ea",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from langchain_community.vectorstores import PathwayVectorClient"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "51d365aa-5526-4ec8-932d-e65e83ee355d",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"vectorstore_client = PathwayVectorClient(pathway_host, pathway_port)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "59624377-8481-4407-8ee3-0f014bdff30e",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"# this will not be able to fetch data until the RAG app is running\n",
"retriever = vectorstore_client.as_retriever()"
]
},
{
"attachments": {
"76895b7a-fcc5-4758-9fbb-510b17fdeda4.png": {
"image/png": "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"
}
},
"cell_type": "markdown",
"id": "5894115c-0323-4098-83be-da49d949a53c",
"metadata": {
"tags": []
},
"source": [
"### Self-RAG\n",
"\n",
"Self-RAG is a strategy for RAG that incorporates self-reflection / self-grading on retrieved documents and generations.\n",
"\n",
"In the [paper](https://arxiv.org/abs/2310.11511), a few decisions are made:\n",
"\n",
"1. Should I retrieve from retriever, `R` -\n",
"\n",
"* Input: `x (question)` OR `x (question)`, `y (generation)`\n",
"* Decides when to retrieve `D` chunks with `R`\n",
"* Output: `yes, no, continue`\n",
"\n",
"2. Are the retrieved passages `D` relevant to the question `x` -\n",
"\n",
"* * Input: (`x (question)`, `d (chunk)`) for `d` in `D`\n",
"* `d` provides useful information to solve `x`\n",
"* Output: `relevant, irrelevant`\n",
"\n",
"3. Are the LLM generation from each chunk in `D` is relevant to the chunk (hallucinations, etc) -\n",
"\n",
"* Input: `x (question)`, `d (chunk)`, `y (generation)` for `d` in `D`\n",
"* All of the verification-worthy statements in `y (generation)` are supported by `d`\n",
"* Output: `{fully supported, partially supported, no support`\n",
"\n",
"4. The LLM generation from each chunk in `D` is a useful response to `x (question)` -\n",
"\n",
"* Input: `x (question)`, `y (generation)` for `d` in `D`\n",
"* `y (generation)` is a useful response to `x (question)`.\n",
"* Output: `{5, 4, 3, 2, 1}`\n",
"\n",
"We will implement some of these ideas from scratch using [LangGraph](https://langchain-ai.github.io/langgraph/).\n",
"\n",
"> Figure and the explanation taken from the [LangGraph cookbook](https://github.com/langchain-ai/langgraph/blob/e3ca7bb3e9d34b09633852f4d08d55f6dcd4364b/examples/rag/langgraph_self_rag.ipynb)"
]
},
{
"cell_type": "markdown",
"id": "1def042d-abcc-47af-8926-a9d1e2e8154f",
"metadata": {},
"source": [
"### Define the LangGraph Workflow\n",
"\n",
"\n",
"The LangGraph agent operates as a multi-step workflow.\n",
"\n",
"- It retrieves documents, grades their relevance, generates an answer, and ensures factual accuracy (e.g., no hallucinations).\n",
"- You can also implement self-reflection and query optimization using LangGraph nodes."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8c4c9be8-396c-4ebc-a9e4-d4e11504c05c",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from langchain_core.prompts import ChatPromptTemplate\n",
"from langchain_core.pydantic_v1 import BaseModel, Field\n",
"from langchain_openai import ChatOpenAI\n",
"\n",
"\n",
"# Data model\n",
"class GradeDocuments(BaseModel):\n",
" \"\"\"Binary score for relevance check on retrieved documents.\"\"\"\n",
"\n",
" binary_score: str = Field(\n",
" description=\"Documents are relevant to the question, 'yes' or 'no'\"\n",
" )\n",
"\n",
"\n",
"# LLM with function call\n",
"llm = ChatOpenAI(model=\"gpt-4o-mini\", temperature=0)\n",
"structured_llm_grader = llm.with_structured_output(GradeDocuments)\n",
"\n",
"# Prompt\n",
"system = \"\"\"You are a grader assessing relevance of a retrieved document to a user question. \\n \n",
" It does not need to be a stringent test. The goal is to filter out erroneous retrievals. \\n\n",
" If the document contains keyword(s) or semantic meaning related to the user question, grade it as relevant. \\n\n",
" Give a binary score 'yes' or 'no' score to indicate whether the document is relevant to the question.\"\"\"\n",
"grade_prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", system),\n",
" (\"human\", \"Retrieved document: \\n\\n {document} \\n\\n User question: {question}\"),\n",
" ]\n",
")\n",
"\n",
"retrieval_grader = grade_prompt | structured_llm_grader"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "499173a5-8df0-4fe6-bb46-552ce496b8ba",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from langchain import hub\n",
"from langchain_core.output_parsers import StrOutputParser\n",
"\n",
"# Prompt\n",
"prompt = hub.pull(\"rlm/rag-prompt\")\n",
"\n",
"# LLM\n",
"llm = ChatOpenAI(model_name=\"gpt-3.5-turbo\", temperature=0)\n",
"\n",
"\n",
"# Post-processing\n",
"def format_docs(docs):\n",
" return \"\\n\\n\".join(doc.page_content for doc in docs)\n",
"\n",
"\n",
"# Chain\n",
"rag_chain = prompt | llm | StrOutputParser()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "d9055d3b-8e9b-494f-bf4f-d02c153639c7",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"class GradeHallucinations(BaseModel):\n",
" \"\"\"Binary score for hallucination present in generation answer.\"\"\"\n",
"\n",
" binary_score: str = Field(\n",
" description=\"Answer is grounded in the facts, 'yes' or 'no'\"\n",
" )\n",
"\n",
"\n",
"# LLM with function call\n",
"llm = ChatOpenAI(model=\"gpt-4o-mini\", temperature=0)\n",
"structured_llm_grader = llm.with_structured_output(GradeHallucinations)\n",
"\n",
"# Prompt\n",
"system = \"\"\"You are a grader assessing whether an LLM generation is grounded in / supported by a set of retrieved facts. \\n \n",
" Give a binary score 'yes' or 'no'. 'Yes' means that the answer is grounded in / supported by the set of facts.\"\"\"\n",
"hallucination_prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", system),\n",
" (\"human\", \"Set of facts: \\n\\n {documents} \\n\\n LLM generation: {generation}\"),\n",
" ]\n",
")\n",
"\n",
"hallucination_grader = hallucination_prompt | structured_llm_grader"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "f92d351d-5ee1-48cc-ad8f-9b57ceed18ed",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"class GradeAnswer(BaseModel):\n",
" \"\"\"Binary score to assess answer addresses question.\"\"\"\n",
"\n",
" binary_score: str = Field(\n",
" description=\"Answer addresses the question, 'yes' or 'no'\"\n",
" )\n",
"\n",
"\n",
"# LLM with function call\n",
"llm = ChatOpenAI(model=\"gpt-4o-mini\", temperature=0)\n",
"structured_llm_grader = llm.with_structured_output(GradeAnswer)\n",
"\n",
"# Prompt\n",
"system = \"\"\"You are a grader assessing whether an answer addresses / resolves a question \\n \n",
" Give a binary score 'yes' or 'no'. Yes' means that the answer resolves the question.\"\"\"\n",
"answer_prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", system),\n",
" (\"human\", \"User question: \\n\\n {question} \\n\\n LLM generation: {generation}\"),\n",
" ]\n",
")\n",
"\n",
"answer_grader = answer_prompt | structured_llm_grader"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "286a9a2a-2a62-4004-84a2-fc82e6996cf1",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"llm = ChatOpenAI(model=\"gpt-4o-mini\", temperature=0)\n",
"\n",
"# Prompt\n",
"system = \"\"\"You a question re-writer that converts an input question to a better version that is optimized \\n \n",
" for vectorstore retrieval. Look at the input and try to reason about the underlying semantic intent / meaning.\"\"\"\n",
"re_write_prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", system),\n",
" (\n",
" \"human\",\n",
" \"Here is the initial question: \\n\\n {question} \\n Formulate an improved question.\",\n",
" ),\n",
" ]\n",
")\n",
"\n",
"question_rewriter = re_write_prompt | llm | StrOutputParser()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "cb9bd597-a683-433a-879d-7991cfed5cf7",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from typing import List\n",
"\n",
"from typing_extensions import TypedDict\n",
"\n",
"\n",
"class GraphState(TypedDict):\n",
" \"\"\"\n",
" Represents the state of our graph.\n",
"\n",
" Attributes:\n",
" question: question\n",
" generation: LLM generation\n",
" documents: list of documents\n",
" \"\"\"\n",
"\n",
" question: str\n",
" generation: str\n",
" documents: List[str]"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "398e5cc2-9c41-465d-a4f4-93c82ed452f2",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"def retrieve(state):\n",
" \"\"\"\n",
" Retrieve documents\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" state (dict): New key added to state, documents, that contains retrieved documents\n",
" \"\"\"\n",
" print(\"---RETRIEVE---\")\n",
" question = state[\"question\"]\n",
"\n",
" # Retrieval\n",
" documents = retriever.get_relevant_documents(question)\n",
" return {\"documents\": documents, \"question\": question}\n",
"\n",
"\n",
"def generate(state):\n",
" \"\"\"\n",
" Generate answer\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" state (dict): New key added to state, generation, that contains LLM generation\n",
" \"\"\"\n",
" print(\"---GENERATE---\")\n",
" question = state[\"question\"]\n",
" documents = state[\"documents\"]\n",
"\n",
" # RAG generation\n",
" generation = rag_chain.invoke({\"context\": documents, \"question\": question})\n",
" return {\"documents\": documents, \"question\": question, \"generation\": generation}\n",
"\n",
"\n",
"def grade_documents(state):\n",
" \"\"\"\n",
" Determines whether the retrieved documents are relevant to the question.\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" state (dict): Updates documents key with only filtered relevant documents\n",
" \"\"\"\n",
"\n",
" print(\"---CHECK DOCUMENT RELEVANCE TO QUESTION---\")\n",
" question = state[\"question\"]\n",
" documents = state[\"documents\"]\n",
"\n",
" # Score each doc\n",
" filtered_docs = []\n",
" for d in documents:\n",
" score = retrieval_grader.invoke(\n",
" {\"question\": question, \"document\": d.page_content}\n",
" )\n",
" grade = score.binary_score\n",
" if grade == \"yes\":\n",
" print(\"---GRADE: DOCUMENT RELEVANT---\")\n",
" filtered_docs.append(d)\n",
" else:\n",
" print(\"---GRADE: DOCUMENT NOT RELEVANT---\")\n",
" continue\n",
" return {\"documents\": filtered_docs, \"question\": question}\n",
"\n",
"\n",
"def transform_query(state):\n",
" \"\"\"\n",
" Transform the query to produce a better question.\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" state (dict): Updates question key with a re-phrased question\n",
" \"\"\"\n",
"\n",
" print(\"---TRANSFORM QUERY---\")\n",
" question = state[\"question\"]\n",
" documents = state[\"documents\"]\n",
"\n",
" # Re-write question\n",
" better_question = question_rewriter.invoke({\"question\": question})\n",
" return {\"documents\": documents, \"question\": better_question}\n",
"\n",
"\n",
"### Edges\n",
"\n",
"\n",
"def decide_to_generate(state):\n",
" \"\"\"\n",
" Determines whether to generate an answer, or re-generate a question.\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" str: Binary decision for next node to call\n",
" \"\"\"\n",
"\n",
" print(\"---ASSESS GRADED DOCUMENTS---\")\n",
" state[\"question\"]\n",
" filtered_documents = state[\"documents\"]\n",
"\n",
" if not filtered_documents:\n",
" # All documents have been filtered check_relevance\n",
" # We will re-generate a new query\n",
" print(\n",
" \"---DECISION: ALL DOCUMENTS ARE NOT RELEVANT TO QUESTION, TRANSFORM QUERY---\"\n",
" )\n",
" return \"transform_query\"\n",
" else:\n",
" # We have relevant documents, so generate answer\n",
" print(\"---DECISION: GENERATE---\")\n",
" return \"generate\"\n",
"\n",
"\n",
"def grade_generation_v_documents_and_question(state):\n",
" \"\"\"\n",
" Determines whether the generation is grounded in the document and answers question.\n",
"\n",
" Args:\n",
" state (dict): The current graph state\n",
"\n",
" Returns:\n",
" str: Decision for next node to call\n",
" \"\"\"\n",
"\n",
" print(\"---CHECK HALLUCINATIONS---\")\n",
" question = state[\"question\"]\n",
" documents = state[\"documents\"]\n",
" generation = state[\"generation\"]\n",
"\n",
" score = hallucination_grader.invoke(\n",
" {\"documents\": documents, \"generation\": generation}\n",
" )\n",
" grade = score.binary_score\n",
"\n",
" # Check hallucination\n",
" if grade == \"yes\":\n",
" print(\"---DECISION: GENERATION IS GROUNDED IN DOCUMENTS---\")\n",
" # Check question-answering\n",
" print(\"---GRADE GENERATION vs QUESTION---\")\n",
" score = answer_grader.invoke({\"question\": question, \"generation\": generation})\n",
" grade = score.binary_score\n",
" if grade == \"yes\":\n",
" print(\"---DECISION: GENERATION ADDRESSES QUESTION---\")\n",
" return \"useful\"\n",
" else:\n",
" print(\"---DECISION: GENERATION DOES NOT ADDRESS QUESTION---\")\n",
" return \"not useful\"\n",
" else:\n",
" pprint(\"---DECISION: GENERATION IS NOT GROUNDED IN DOCUMENTS, RE-TRY---\")\n",
" return \"not supported\""
]
},
{
"cell_type": "markdown",
"id": "b77061d5-fa7f-4dbc-ad5b-cc409b931bf4",
"metadata": {},
"source": [
"### Compose the graph"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "f5ae35f7-ff5c-40f1-9b0e-1a8bac0dc58f",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from langgraph.graph import END, StateGraph, START\n",
"\n",
"workflow = StateGraph(GraphState)\n",
"\n",
"# Define the nodes\n",
"workflow.add_node(\"retrieve\", retrieve) # retrieve\n",
"workflow.add_node(\"grade_documents\", grade_documents) # grade documents\n",
"workflow.add_node(\"generate\", generate) # generatae\n",
"workflow.add_node(\"transform_query\", transform_query) # transform_query\n",
"\n",
"# Build graph\n",
"workflow.add_edge(START, \"retrieve\")\n",
"workflow.add_edge(\"retrieve\", \"grade_documents\")\n",
"workflow.add_conditional_edges(\n",
" \"grade_documents\",\n",
" decide_to_generate,\n",
" {\n",
" \"transform_query\": \"transform_query\",\n",
" \"generate\": \"generate\",\n",
" },\n",
")\n",
"workflow.add_edge(\"transform_query\", \"retrieve\")\n",
"workflow.add_conditional_edges(\n",
" \"generate\",\n",
" grade_generation_v_documents_and_question,\n",
" {\n",
" \"not supported\": \"generate\",\n",
" \"useful\": END,\n",
" \"not useful\": \"transform_query\",\n",
" },\n",
")\n",
"\n",
"# Compile\n",
"app = workflow.compile()"
]
},
{
"cell_type": "markdown",
"id": "598a9ef4-cfb2-479a-9a74-b82c5781daca",
"metadata": {},
"source": [
"# Serve the agent\n",
"\n",
"`BaseRAGQuestionAnswerer` provides `serve_callable` that lets you attach a custom Python function to an endpoint. This way, you can create your own endpoints on the Pathway server in addition to the [built-in ones](https://pathway.com/developers/ai-pipelines/rest-api).\n",
"\n",
"Use the `serve_callable` method to expose the LangGraph agent as an HTTP endpoint.\n",
"The `/agent` endpoint accepts user queries, processes them through the LangGraph pipeline, and returns the final answer."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "fafc447a-c221-4251-ad40-c56cb77c338c",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"@rag_app.serve_callable(route=\"/agent\")\n",
"async def call_agent(user_query: str) -> str:\n",
" langgraph_input = {\"question\": user_query}\n",
" result = app.invoke(langgraph_input)\n",
" return result[\"generation\"]"
]
},
{
"cell_type": "markdown",
"id": "a71d3943-faf8-4f74-9f9d-a8f914db2acb",
"metadata": {},
"source": [
"The above Pathway endpoint,\n",
"* Takes a `user_query` as the JSON input.\n",
"* Passes it into the compiled LangGraph pipeline.\n",
"* Returns the final answer from the state (result[\"generation\"])"
]
},
{
"cell_type": "markdown",
"id": "6e86ac7c-db40-4636-b704-77df7c0d18a6",
"metadata": {},
"source": [
"# Build and Run the Pathway server"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "963e891e-4ccf-4186-a3f2-19997b44bb86",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"# This creates and runs the index, also serves the agent endpoint defined above\n",
"\n",
"rag_app.build_server(pathway_host, pathway_port)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "065f68f6-99b9-4d5e-8abf-796fdf67c050",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"t = rag_app.run_server(threaded=True)"
]
},
{
"cell_type": "markdown",
"id": "95b9772b-42ce-422f-bd40-56ba317a8249",
"metadata": {},
"source": [
"List the indexed documents.\n",
"\n",
"This request may take a while since app is parsing the documents and building the index."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8bc8ab3c-cfb6-41e6-a102-281caa1bb19e",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"pathway_client = RAGClient(pathway_host, pathway_port)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "aebfab0c-c827-4a47-a5b4-36361c2c68c1",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"pathway_client.list_documents()"
]
},
{
"cell_type": "markdown",
"id": "1e682805-8d61-4054-aab0-47044d921157",
"metadata": {},
"source": [
"# Run with a sample question\n",
"\n",
"Test the `/agent` endpoint by sending a sample question. The server runs the entire RAG pipeline and returns the answer."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "4676f4c8-26bd-4f36-aadf-80771b001b59",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"question: str = (\n",
" \"How does self-RAG work? Explain the steps involved in the implementation.\"\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "b84144d8-e266-4a13-a515-a55fcdc02f11",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import requests\n",
"\n",
"# send request to the endpoint we defined above, put the expected fields as payload\n",
"response = requests.post(\n",
" pathway_client.url + \"/agent\",\n",
" json={\"user_query\": question},\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a5689309-1780-4725-8d3f-364841303def",
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"response.json()"
]
},
{
"cell_type": "markdown",
"id": "35051a4c-1c70-4631-9005-5a06520640e1",
"metadata": {
"tags": []
},
"source": [
"## Summary\n",
"\n",
"1. **Data Ingestion:** Ingest a webpage (the Self-RAG arxiv page).\n",
"2. **Indexing:** Pathway reads these text documents, parses, splits, and indexes them using a hybrid BM25 + semantic index.\n",
"3. **LangGraph Agent:** Define a multi-step agent that:\n",
" * Re-writes the query if needed.\n",
" * Retrieves documents via the Pathway VectorStore.\n",
" * Grades those docs for relevance.\n",
" * Generates an answer.\n",
" * Grades that answer for hallucination and for how well it addresses the user’s question.\n",
" * Loops as necessary until it gets a satisfactory final response.\n",
"4. **Serving:** Pathway’s built-in server is launched, hosting an `/agent` endpoint for queries.\n",
"\n",
"\n",
"Thus, whenever you POST a `\"user_query\"` to the `/agent` endpoint, the agent pipeline runs, and returns the final answer."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.11"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
```
## /pyproject.toml
```toml path="/pyproject.toml"
[tool.poetry]
name = "llm-app"
version = "0.3.6"
description = "LLM-App is a library for creating responsive AI applications leveraging OpenAI/Hugging Face APIs to provide responses to user queries based on live data sources. Build your own LLM application in 30 lines of code, no vector database required."
authors = [
"Jan Chorowski <janek@pathway.com>",
"Kamil Piechowiak <kamil@pathway.com>",
"Jakub Kowalski <kuba@pathway.com>",
"Michał Bartoszkiewicz <embe@pathway.com>",
"Mohamed Malhou <mohamed@pathway.com>",
"Sergey Kulik <sergey@pathway.com>",
"Adrian Kosowski <adrian@pathway.com>",
"Mateusz Lewandowski <mateusz@pathway.com>",
"Olivier Ruas <olivier@pathway.com>",
]
license = "MIT"
readme = "README.md"
keywords = ["Pathway", "LLM"]
package-mode = false
classifiers = [
"Development Status :: 3 - Alpha",
"Intended Audience :: Developers",
"Programming Language :: Python :: Implementation :: CPython",
"License :: OSI Approved :: MIT License",
]
[tool.poetry.urls]
"Homepage" = "https://github.com/pathwaycom/llm-app"
"Source Code" = "https://github.com/pathwaycom/llm-app"
[tool.poetry.dependencies]
python = ">=3.10,<3.13"
pathway = "^0.12.0"
[tool.poetry.group.linters]
optional = true
[tool.poetry.group.linters.dependencies]
black = "^24.2.0"
isort = "^5.13.2"
mypy = "~1.10.0"
flake8 = "~7.0.0"
pytest = "^8.0.2"
types-requests = "^2.31.0"
types-PyYAML = "^6.0.0"
[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"
[tool.black]
target-version = ["py310", "py311"]
[tool.isort]
profile = "black"
[tool.mypy]
python_version = "3.10"
exclude = ["templates/adaptive_rag", "templates/private_rag", "templates/document_indexing", "templates/multimodal_rag", "templates/question_answering_rag"]
ignore_missing_imports = true
check_untyped_defs = true
warn_redundant_casts = true
warn_unused_ignores = true
strict_equality = true
```
## /setup.cfg
```cfg path="/setup.cfg"
[flake8]
exclude =
.git
max-line-length = 119
docstring-convention = google
extend-ignore =
# Black (https://black.readthedocs.io/en/stable/guides/using_black_with_other_tools.html#flake8)
E203, E701
```
## /templates/adaptive_rag/.env.example
```example path="/templates/adaptive_rag/.env.example"
OPENAI_API_KEY="sk-***"
```
## /templates/adaptive_rag/Dockerfile
``` path="/templates/adaptive_rag/Dockerfile"
FROM pathwaycom/pathway:latest
WORKDIR /app
RUN apt-get update \
&& apt-get install -y python3-opencv tesseract-ocr-eng \
&& rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/*
COPY requirements.txt .
RUN pip install -U --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8000
CMD ["python", "app.py"]
```
## /templates/adaptive_rag/app.py
```py path="/templates/adaptive_rag/app.py"
import logging
from warnings import warn
import pathway as pw
from dotenv import load_dotenv
from pathway.xpacks.llm.question_answering import SummaryQuestionAnswerer
from pathway.xpacks.llm.servers import QASummaryRestServer
from pydantic import BaseModel, ConfigDict, InstanceOf
# To use advanced features with Pathway Scale, get your free license key from
# https://pathway.com/features and paste it below.
# To use Pathway Community, comment out the line below.
pw.set_license_key("demo-license-key-with-telemetry")
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s %(name)s %(levelname)s %(message)s",
datefmt="%Y-%m-%d %H:%M:%S",
)
load_dotenv()
class App(BaseModel):
question_answerer: InstanceOf[SummaryQuestionAnswerer]
host: str = "0.0.0.0"
port: int = 8000
with_cache: bool | None = None # deprecated
persistence_backend: pw.persistence.Backend | None = None
persistence_mode: pw.PersistenceMode | None = pw.PersistenceMode.UDF_CACHING
terminate_on_error: bool = False
def run(self) -> None:
server = QASummaryRestServer( # noqa: F841
self.host, self.port, self.question_answerer
)
if self.persistence_mode is None:
if self.with_cache is True:
warn(
"`with_cache` is deprecated. Please use `persistence_mode` instead.",
DeprecationWarning,
)
persistence_mode = pw.PersistenceMode.UDF_CACHING
else:
persistence_mode = None
else:
persistence_mode = self.persistence_mode
if persistence_mode is not None:
if self.persistence_backend is None:
persistence_backend = pw.persistence.Backend.filesystem("./Cache")
else:
persistence_backend = self.persistence_backend
persistence_config = pw.persistence.Config(
persistence_backend,
persistence_mode=persistence_mode,
)
else:
persistence_config = None
pw.run(
persistence_config=persistence_config,
terminate_on_error=self.terminate_on_error,
monitoring_level=pw.MonitoringLevel.NONE,
)
model_config = ConfigDict(extra="forbid", arbitrary_types_allowed=True)
if __name__ == "__main__":
with open("app.yaml") as f:
config = pw.load_yaml(f)
app = App(**config)
app.run()
```
## /templates/adaptive_rag/app.yaml
```yaml path="/templates/adaptive_rag/app.yaml"
# This YAML configuration file is used to set up and configure the Adaptive RAG template.
# It defines various components such as data sources, language models, embedders, splitters, parsers, and retrievers.
# Each section is configured to specify how the template should process and handle data for generating responses.
# You can learn more about the YAML syntax here: https://pathway.com/developers/templates/configure-yaml
# $sources defines the data sources used to read the data which will be indexed in the RAG.
# You can learn more how to configure data sources here:
# https://pathway.com/developers/templates/yaml-examples/data-sources-examples
$sources:
# File System connector, reading data locally.
- !pw.io.fs.read
path: data
format: binary
with_metadata: true
# Uncomment to use the SharePoint connector
# - !pw.xpacks.connectors.sharepoint.read
# url: $SHAREPOINT_URL
# tenant: $SHAREPOINT_TENANT
# client_id: $SHAREPOINT_CLIENT_ID
# cert_path: sharepointcert.pem
# thumbprint: $SHAREPOINT_THUMBPRINT
# root_path: $SHAREPOINT_ROOT
# with_metadata: true
# refresh_interval: 30
# Uncomment to use the Google Drive connector
# - !pw.io.gdrive.read
# object_id: $DRIVE_ID
# service_user_credentials_file: gdrive_indexer.json
# file_name_pattern:
# - "*.pdf"
# - "*.pptx"
# object_size_limit: null
# with_metadata: true
# refresh_interval: 30
# Configures the LLM model settings for generating responses.
# The list of available Pathway LLM wrappers is available here:
# https://pathway.com/developers/api-docs/pathway-xpacks-llm/llms
# You can learn more about those in our documentation:
# https://pathway.com/developers/templates/rag-customization/llm-chats
$llm: !pw.xpacks.llm.llms.OpenAIChat
model: "gpt-4.1-mini"
retry_strategy: !pw.udfs.ExponentialBackoffRetryStrategy
max_retries: 6
cache_strategy: !pw.udfs.DefaultCache {}
temperature: 0
capacity: 8
# Specifies the embedder model for converting text into embeddings.
$embedder: !pw.xpacks.llm.embedders.OpenAIEmbedder
model: "text-embedding-3-small"
cache_strategy: !pw.udfs.DefaultCache {}
retry_strategy: !pw.udfs.ExponentialBackoffRetryStrategy {}
# Defines the splitter settings for dividing text into smaller chunks.
$splitter: !pw.xpacks.llm.splitters.TokenCountSplitter
max_tokens: 400
# Configures the parser for processing and extracting information from documents.
$parser: !pw.xpacks.llm.parsers.DoclingParser
async_mode: "fully_async"
chunk: false
cache_strategy: !pw.udfs.DefaultCache {}
# Sets up the retriever factory for indexing and retrieving documents.
$retriever_factory: !pw.indexing.UsearchKnnFactory
reserved_space: 1000
embedder: $embedder
metric: !pw.indexing.USearchMetricKind.COS
# Manages the storage and retrieval of documents for the RAG template.
$document_store: !pw.xpacks.llm.document_store.DocumentStore
docs: $sources
parser: $parser
splitter: $splitter
retriever_factory: $retriever_factory
# Configures the question-answering component using the RAG approach.
# The component builds a RAG over an index.
# You can interact with obtained RAG using a REST API.
# You can learn more about the available operations here:
# https://pathway.com/developers/templates/rag-customization/rest-api
question_answerer: !pw.xpacks.llm.question_answering.AdaptiveRAGQuestionAnswerer
llm: $llm
indexer: $document_store
n_starting_documents: 2
factor: 2
max_iterations: 4
# Change host and port of the webserver by uncommenting these lines
# host: "0.0.0.0"
# port: 8000
# By default, caching is enabled for UDFs with cache_strategy set.
# You can disable it by uncommenting the following line.
# persistence_mode: null
# You can also set persistence_mode to !pw.PersistenceMode.PERSISTING to enable persistence
# across restarts.
# By default, when enabled, Cache is stored in .Cache directory.
# You can customize the location by uncommenting and modifying the following lines:
# persistence_backend: !pw.persistence.Backend.filesystem
# path: ".Cache"
# If `terminate_on_error` is true then the program will terminate whenever any error is encountered.
# Defaults to false, uncomment the following line if you want to set it to true
# terminate_on_error: true
```
## /templates/adaptive_rag/data/IdeanomicsInc_20160330_10-K_EX-10.26_9512211_EX-10.26_Content License Agreement.pdf
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/templates/adaptive_rag/data/IdeanomicsInc_20160330_10-K_EX-10.26_9512211_EX-10.26_Content License Agreement.pdf
## /templates/adaptive_rag/requirements.txt
python-dotenv~=1.0
## /templates/document_indexing/files-for-indexing/2023q2-alphabet-earnings-release.pdf
Binary file available at https://raw.githubusercontent.com/pathwaycom/llm-app/refs/heads/main/templates/document_indexing/files-for-indexing/2023q2-alphabet-earnings-release.pdf
The content has been capped at 50000 tokens. The user could consider applying other filters to refine the result. The better and more specific the context, the better the LLM can follow instructions. If the context seems verbose, the user can refine the filter using uithub. Thank you for using https://uithub.com - Perfect LLM context for any GitHub repo.