infiniflow/ragflow
RAGFlow is an open-source Retrieval-Augmented Generation engine that leverages deep document understanding and Large Language Models to provide accurate, citation-backed question-answering from complex and diverse data sources.
Awesome AI Agents › NL AI Frameworks
HumanSignal/Adala
Adala is an Autonomous Data Labeling Agent framework designed to facilitate the creation and deployment of intelligent agents specialized in data processing and labeling tasks. The framework enables agents to autonomously acquire and refine skills through iterative learning influenced by their environment, observations, and reflections. Users provide a ground truth dataset that defines the environment, allowing agents to learn and apply their skills within a runtime environment, typically powered by Large Language Models (LLMs) such as OpenAI's GPT models or VertexAI. Adala emphasizes reliability by grounding agent learning in verified data, ensuring consistent and trustworthy outputs. It offers controllable output configurations, allowing users to specify constraints and desired results with varying flexibility, from strict adherence to guidelines to adaptive learning-based outputs. The framework is highly specialized for data processing but is flexible enough to be customized for a wide range of tasks. The runtime environment in Adala is extensible and adaptable, supporting deployment of a single skill across multiple runtimes and enabling complex architectures like student/teacher models. This openness encourages community contributions and continuous evolution. Adala is accessible via command line, RESTful API, and direct integration in Python notebooks or scripts, making it versatile for different user preferences. Target users include AI engineers, machine learning researchers, data scientists, educators, and students. AI engineers can build modular agent systems abstracting low-level machine learning details, researchers can experiment with problem decomposition and causal reasoning, data scientists can preprocess and postprocess data efficiently, and educators can use Adala as a teaching tool or research base. Installation is straightforward via pip, with options for stable releases or the latest development version from GitHub. The framework requires an API key for LLM providers like OpenAI. Adala supports integration with tools like Label Studio for human-in-the-loop labeling and ground truth dataset creation. Overall, Adala provides a robust, flexible, and intelligent framework for autonomous data labeling and processing agents, leveraging modern LLM capabilities to enhance AI workflows and research.
https://github.com/HumanSignal/Adala
Also in NL AI Frameworks
ScrapeGraphAI/Scrapegraph-ai
ScrapeGraphAI is an AI-powered Python library that creates intelligent web scraping pipelines using large language models to extract structured data from websites and local documents efficiently.
pydantic/pydantic
Pydantic is a Python library for fast and extensible data validation using Python type hints, enabling developers to define and validate data models efficiently.
apache/doris
Apache Doris is a high-performance, real-time analytical database with a storage-compute integrated architecture, designed for sub-second query response and high concurrency in diverse data analysis scenarios.
business-science/ai-data-science-team
AI Data Science Team is a Python library featuring AI-powered agents that automate and accelerate common data science tasks, including data cleaning, feature engineering, machine learning, and exploratory data analysis, to improve productivity and efficiency.
ucbepic/docetl
DocETL is a system for creating and executing complex document processing and data transformation pipelines powered by large language models, featuring an interactive UI playground and a Python package for production use.
hitsz-ids/synthetic-data-generator
Synthetic Data Generator (SDG) is a comprehensive framework for generating high-quality, privacy-compliant synthetic structured tabular data using advanced statistical and LLM-based models, optimized for big data applications.
Mintplex-Labs/vector-admin
VectorAdmin is a universal and user-friendly tool suite for managing multiple vector databases, providing full control over vector data with multi-user support, embedding management, and cloud deployment capabilities.