KET-RAG

KET-RAG is a retrieval-augmented generation framework that incorporates knowledge graphs for efficient document indexing and answer generation.

CommonProductProgrammingKnowledge GraphRetrieval-Augmented Generation

Visit

KET-RAG (Knowledge-Enhanced Text Retrieval Augmented Generation) is a powerful retrieval-augmented generation framework enhanced with knowledge graph technology. It achieves efficient knowledge retrieval and generation through a multi-granularity indexing framework, such as a knowledge graph skeleton and a text-keyword bipartite graph. This framework significantly improves retrieval and generation quality while reducing indexing costs, making it well-suited for large-scale RAG applications. Developed in Python, KET-RAG supports flexible configuration and extension, catering to the needs of developers and researchers seeking efficient knowledge retrieval and generation.

Supports Knowledge Graph Skeleton (SkeletonRAG) to select key text fragments and extract structured knowledge.
Efficiently links keywords to text fragments through a Text-Keyword Bipartite Graph (KeywordRAG).
Combines entity and keyword channels for efficient retrieval and high-quality generation.
Supports dependency installation via Poetry
facilitating environment configuration and management.
Provides flexible indexing and context generation tools suitable for a variety of applications.

KET-RAG is designed for developers
researchers
and enterprise application developers who require efficient knowledge retrieval and generation. It enables users to quickly retrieve relevant information from large-scale documents and generate high-quality answers
making it suitable for question answering systems
intelligent customer service

In question answering systems
KET-RAG can quickly retrieve knowledge base entries and generate accurate answers.
For intelligent customer service scenarios
KET-RAG can retrieve relevant knowledge based on user questions and generate responses.
In knowledge management systems

1. **Install Dependencies:** Use Poetry to install project dependencies.
2. **Initialize Project:** Run the initialization command to set up the project's file structure.
3. **Tune Prompts:** Optimize retrieval performance by adjusting prompts using the `prompt-tune` command.
4. **Build Index:** Create knowledge graph and text indexes by running the indexing command.
5. **Generate Context and Answers:** Use the `create_context.py` and `llm_answer.py` scripts to generate context and answers.

Visit

KET-RAG Visit Over Time

Monthly Visits

502571820

Bounce Rate

37.10%

Page per Visit

5.9

Visit Duration

00:06:29

KET-RAG Visit Trend

KET-RAG Visit Geography

KET-RAG Traffic Sources

KET-RAG Alternatives

iText2KG — Incrementally build knowledge graphs using large language models.

Productivity•Knowledge Graph•Natural Language Processing

318

MiniRAG — A simple retrieval-augmented generation framework that enables small models to achieve good RAG performance through heterogeneous graph indexing and lightweight topological enhanced retrieval.

Programming•Retrieval-Augmented Generation•Small Language Models

Data Formulator — AI-powered data visualization tool

Productivity•Data Visualization•Natural Language Processing

390

Knowledge Table — An open-source tool that simplifies the extraction and exploration of structured data from unstructured documents.

Programming•Natural Language Processing•Data Extraction

372

LightRAG — A simple and fast retrieval-augmented generation model

Programming•Natural Language Processing•Retrieval-Augmented Generation

174

LlamaIndex.TS — A framework for building LLM applications

Programming•LLM•Natural Language Processing

C4AI Command R+ 08-2024 — A multilingual advanced dialogue generation model with 104 billion parameters.

Productivity•Artificial Intelligence•Machine Learning

fact-finder — An intelligent question-answering system based on knowledge graphs.

Programming•Intelligent Q&A•Knowledge Graph

174

Easy-RAG — A RAG system suitable for learning, use, and self-extension.

Programming•RAG•Retrieval-Augmented Generation

1038

Datalore — AI-driven data analytics tool

Programming•Data Analytics•Natural Language Processing

216

Knowledge Graph RAG — Enhances the performance of language models using knowledge graphs and document networks

Programming•Knowledge Graph•Document Network

684

Prettygraph — An experimental UI that transforms text into knowledge graphs.

Programming•Knowledge Graph•Text Analysis

234

GraphRAG — An end-to-end system that integrates text extraction, network analysis, and large language model prompting and summarization

Programming•Artificial Intelligence•Natural Language Processing

2652

MindGraph — A concept proof for an AI-generated and continuously expanding knowledge graph.

Productivity•Knowledge Graph•Natural Language Processing

258

KG_RAG — A knowledge graph augmented retrieval-generation framework that empowers large language models to handle knowledge-intensive tasks.

Productivity•Knowledge Graph•Natural Language Processing

414

NLTK — Python natural language processing toolkit

Programming•Natural Language Processing•Python

174

Docker GenAI Stack — An AI application development solution launched by Docker

Programming•Docker•Machine Learning

246

Kie.ai — Integrates DeepSeek R1 and V3 APIs on Kie.ai, providing secure and scalable AI solutions.

Others•Reasoning•Natural Language Processing

hallucination-leaderboard — A leaderboard for comparing the hallucination rates of large language models when summarizing short documents.

Others•LLM•Hallucination Detection

LangGraph Multi-Agent Supervisor — A Python library for creating hierarchical multi-agent systems based on LangGraph.

Programming•Multi-Agent System•LangGraph

Proxy — Proxy is a fully automated AI assistant designed to handle daily tasks.

InternationalSelection•AI Assistant•Automation

408

DeepSeek Model Compatibility Checker — Checks whether devices can run various sizes of DeepSeek models and provides compatibility predictions.

Others•Deep Learning•Model Deployment

1176

recurrent-pretraining — Pretraining code for large-scale deep recurrent language models, capable of running on 4096 AMD GPUs.

Programming•Deep Learning•Natural Language Processing

Concierge AI — Engage in natural language interactions with your applications to increase efficiency and convenience.

Productivity•Natural Language Processing•Productivity Tools

174

Zyphra — Zyphra is a company focused on artificial intelligence technology, offering chat models and related services.

chatting•Artificial Intelligence•Chatbot

VideoRAG — VideoRAG is a retrieval-augmented generation framework designed for processing videos with extremely long context.

Video•Video Understanding•Retrieval-Augmented

Basedash — An AI-native business intelligence platform that generates data visualizations and dashboards through natural language.

Business•Business Intelligence•Data Visualization

102

s1-32B — s1 is an inference model fine-tuned based on Qwen2.5-32B-Instruct, trained with only 1,000 samples.

Productivity•Text Generation•Inference Model

336

Site RAG — A Chrome extension for asking questions on websites, supporting local execution and vector storage.

Programming•Chrome Extension•Natural Language Processing

Best AI Websites & Tools

KET-RAG

KET-RAG Visit Over Time

KET-RAG Visit Trend

KET-RAG Visit Geography

KET-RAG Traffic Sources

KET-RAG Alternatives

iText2KG — Incrementally build knowledge graphs using large language models.

MiniRAG — A simple retrieval-augmented generation framework that enables small models to achieve good RAG performance through heterogeneous graph indexing and lightweight topological enhanced retrieval.

Data Formulator — AI-powered data visualization tool

Knowledge Table — An open-source tool that simplifies the extraction and exploration of structured data from unstructured documents.

LightRAG — A simple and fast retrieval-augmented generation model

LlamaIndex.TS — A framework for building LLM applications

C4AI Command R+ 08-2024 — A multilingual advanced dialogue generation model with 104 billion parameters.

fact-finder — An intelligent question-answering system based on knowledge graphs.

Easy-RAG — A RAG system suitable for learning, use, and self-extension.

Datalore — AI-driven data analytics tool

Knowledge Graph RAG — Enhances the performance of language models using knowledge graphs and document networks

Prettygraph — An experimental UI that transforms text into knowledge graphs.

GraphRAG — An end-to-end system that integrates text extraction, network analysis, and large language model prompting and summarization

MindGraph — A concept proof for an AI-generated and continuously expanding knowledge graph.

KG_RAG — A knowledge graph augmented retrieval-generation framework that empowers large language models to handle knowledge-intensive tasks.

NLTK — Python natural language processing toolkit

Docker GenAI Stack — An AI application development solution launched by Docker

Kie.ai — Integrates DeepSeek R1 and V3 APIs on Kie.ai, providing secure and scalable AI solutions.

hallucination-leaderboard — A leaderboard for comparing the hallucination rates of large language models when summarizing short documents.

LangGraph Multi-Agent Supervisor — A Python library for creating hierarchical multi-agent systems based on LangGraph.

Proxy — Proxy is a fully automated AI assistant designed to handle daily tasks.

DeepSeek Model Compatibility Checker — Checks whether devices can run various sizes of DeepSeek models and provides compatibility predictions.

recurrent-pretraining — Pretraining code for large-scale deep recurrent language models, capable of running on 4096 AMD GPUs.

Concierge AI — Engage in natural language interactions with your applications to increase efficiency and convenience.

Zyphra — Zyphra is a company focused on artificial intelligence technology, offering chat models and related services.

VideoRAG — VideoRAG is a retrieval-augmented generation framework designed for processing videos with extremely long context.

Basedash — An AI-native business intelligence platform that generates data visualizations and dashboards through natural language.

s1-32B — s1 is an inference model fine-tuned based on Qwen2.5-32B-Instruct, trained with only 1,000 samples.

Site RAG — A Chrome extension for asking questions on websites, supporting local execution and vector storage.