PyPI page
Home page
Author:
Wissam Metawee
License:
MIT
Summary:
A framework that replace traditional RAG pipelines. Ingest any number of documents in multiple workspaces (channels, departments, etc.), index it with BM25, and let the agent search, fetch, and reason over it, exactly like searching the web, but entirely on your machine. No vector store, no embedding needed.
Latest version:
0.2.1
Required dependencies:
apscheduler
|
beautifulsoup4
|
docling
|
fastapi
|
filelock
|
flashrank
|
google-api-core
|
httpx
|
jinja2
|
langchain-anthropic
|
langchain-core
|
langchain-google-genai
|
langchain-ollama
|
langchain-openai
|
langgraph
|
lxml
|
meilisearch-python-sdk
|
onnxruntime
|
openpyxl
|
platformdirs
|
pyfiglet
|
pymupdf
|
pypdf
|
pytesseract
|
python-docx
|
python-multipart
|
python-pptx
|
pywebview
|
rapidocr
|
rich
|
uvicorn
|
watchdog
Optional dependencies:
mypy
|
pytest
|
pytest-asyncio
|
pytest-cov
|
ruff
Downloads last day:
3
Downloads last week:
90
Downloads last month:
568