bytedance/deer-flow
DeerFlow is a community-driven deep research framework that integrates language models with web search, crawling, and Python execution tools to facilitate comprehensive AI-powered research workflows.
Awesome AI Agents › Reasoning Frameworks
VectifyAI/PageIndex
PageIndex is an innovative document indexing and retrieval system designed to enhance reasoning-based retrieval augmented generation (RAG) without relying on traditional vector databases or chunking methods. It addresses the limitations of vector-based RAG, which often depends on semantic similarity rather than true relevance, by introducing a hierarchical tree index that mimics how human experts navigate complex documents. This system builds a structured "table-of-contents" style tree index for long documents, enabling large language models (LLMs) to perform multi-step reasoning and retrieval through tree search, thus improving accuracy and interpretability. The project is particularly suited for professional and lengthy documents such as financial reports, regulatory filings, academic textbooks, legal, and technical manuals that exceed typical LLM context limits. PageIndex's approach is inspired by AlphaGo's tree search strategy, allowing it to reason over document sections rather than relying on approximate vector similarity searches. This results in a transparent and traceable retrieval process that simulates human-like document analysis. Key features include the elimination of vector databases, avoidance of artificial chunking by organizing documents into natural sections, and a retrieval process based on reasoning that is both interpretable and accurate. The system has demonstrated state-of-the-art performance, achieving 98.7% accuracy on FinanceBench, a benchmark for professional document analysis. PageIndex offers multiple deployment options, including self-hosting via the open-source repository and cloud-based services accessible through a web platform, API, and integrations with multi-agent systems. The project also provides practical resources such as notebooks for vectorless RAG and vision-based RAG workflows, enabling users to experiment with OCR-free document understanding and reasoning-native retrieval. Overall, PageIndex represents a significant advancement in document AI by combining hierarchical indexing with reasoning-based retrieval, making it a powerful tool for extracting relevant information from complex, long-form documents in a human-like manner.
https://github.com/VectifyAI/PageIndex
Also in Reasoning Frameworks
guidance-ai/guidance
Guidance is a programming paradigm and library that enables efficient and fine-grained control over large language models to produce structured, high-quality outputs with reduced latency and cost.
friuns2/BlackFriday-GPTs-Prompts
A curated repository of free GPT prompts and jailbreaks across various domains, enabling users to utilize GPT models without a subscription to GPT Plus.
refly-ai/refly
Refly is an open-source AI-native creation engine that combines multi-threaded dialogues, multimodal inputs, knowledge base integration, and AI-powered tools to transform ideas into production-ready interactive content and web applications.
mongodb-developer/GenAI-Showcase
MongoDB's GenAI Showcase is a comprehensive repository offering examples and applications that integrate MongoDB with advanced Generative AI techniques like Retrieval-Augmented Generation and AI Agents to support AI development and innovation.
anysphere/priompt
Priompt is a JSX-based prompting library that uses a priority system to efficiently construct token-limited prompts for language models, inspired by React's component design.
HKUDS/Auto-Deep-Research
Auto-Deep-Research is an open-source, fully automated, and cost-effective personal AI assistant that supports multiple large language models and offers a one-click launch experience as an alternative to OpenAI's Deep Research.
NirDiamant/Controllable-RAG-Agent
Controllable-RAG-Agent is an advanced Retrieval-Augmented Generation solution featuring a deterministic graph-based autonomous agent designed for complex question answering with multi-step reasoning and hallucination prevention.