Tencent/WeKnora
WeKnora is an LLM-powered modular framework for deep document understanding, semantic retrieval, and context-aware question answering using the Retrieval-Augmented Generation paradigm.
Awesome AI Agents › Document Analytics Platforms
genomoncology/FuzzTypes
FuzzTypes is an extension library for Pydantic that introduces a set of autocorrecting annotation types designed to enhance data validation and normalization beyond Pydantic's built-in capabilities. It focuses on transforming raw input data into structured, meaningful, and corrected forms, ensuring that data fields contain 'smart things' rather than just plain strings. The library provides powerful normalization features such as named entity linking and fuzzy matching, which help in correcting and standardizing data inputs automatically. The project offers a variety of base validators and pre-built usable types that can be directly integrated into Pydantic models. Base validators include classes for date and datetime parsing, regex pattern matching, and in-memory or on-disk entity validation with support for exact, alias, fuzzy, and semantic searches. Usable types cover common data formats such as ASCII text conversion, date and time parsing, email extraction, emoji matching (including fuzzy and semantic matching), integer parsing from words or numbers, language code and name resolution, person name parsing, social security number extraction, and ZIP code extraction. FuzzTypes is designed for simplicity and ease of use, allowing developers to define Pydantic models with annotated fields that automatically correct and normalize input data. For example, it can convert Unicode strings to ASCII, extract emails from complex strings, fuzzy match emoji names, parse human names into components, convert natural language time expressions into datetime objects, and more. The library supports installation via PyPI and provides a Google Colab notebook for interactive experimentation. Overall, FuzzTypes enhances data validation workflows by providing intelligent, autocorrecting types that improve data quality and consistency in Python applications using Pydantic.
https://github.com/genomoncology/FuzzTypes
Also in Document Analytics Platforms
StarTrail-org/PixelRAG
Visual retrieval-augmented generation pipeline that renders web pages, PDFs and images to screenshot tiles and retrieves over them with a LoRA-tuned Qwen3-VL embedding model.
georgia-tech-db/evadb
EvaDB is an open-source database system that empowers developers to build AI-powered applications by integrating advanced AI models and optimizations within a SQL-based querying framework for both structured and unstructured data.
loggie-io/loggie
Loggie is a lightweight, cloud-native data transfer agent and aggregator designed for scalable, high-performance log collection, processing, and monitoring in Kubernetes environments.
johnsonhk88/AI-Bank-Statement-Document-Automation-By-LLM-And-Personal-Finanical-Analysis-Prediction
A full-stack pipeline that parses bank statement PDFs with layout detection, OCR and LLMs, then runs CrewAI, LangChain Deep Agents and Hermes agent harnesses over the extracted data for categorisation, analysis and RAG querying.
netboxlabs/pktvisor
pktvisor is a dynamic network observability agent that analyzes high volume network traffic and generates opentelemetry metrics for real-time and centralized monitoring.