retrieval-systems 开源项目推荐（99 个）

SciPhi-AI /R2R

达到 SoTA 水平、可用于生产的 AI 检索系统。具备 RESTful API 的智能体式检索增强生成（RAG）。

feyninc /chonkie

🦛 CHONK docs with Chonkie ✨ — The lightweight ingestion library for fast, efficient and robust RAG pipelines

EdoardoBotta /RQ-VAE-Recommender

[Pytorch] Generative retrieval model using semantic IDs from "Recommender Systems with Generative Retrieval"

HKUDS /CatchMe

“CatchMe：让你的AI代理真正个性化”

coree /awesome-rag

A curated list of retrieval-augmented generation (RAG) in large language models

feyninc /chonkiejs

🦛 CHONK your texts with Chonkie ✨ Type-friendly, light-weight, fast and super-simple chunking library

pillowsofwind /Knowledge-Conflicts-Survey

[EMNLP 2024] The official GitHub repo for the survey paper "Knowledge Conflicts for LLMs: A Survey"

NonameUntitled /tiger

Unofficial implementation of paper "Recommender Systems with Generative Retrieval" (NeurIPS'23)

gmftbyGMFTBY /SimpleReDial-v1

The sources codes of the DR-BERT model and baselines

fredfung007 /cityflow-nl

Tracked Vehicle Retrieval by NL Challenge in the 2023 AI City Challenge.

fanzfeng /QASystem

简版文本对话/问答系统

anakin87 /neural-search-pills

Knowledge pills on Neural Search

sebhaan /SemanticCite

AI-powered citation verification with full-text analysis and contextual insight. Automatically validates citations with evidence-based reasoning, confidence scores, and support classifications. Features fine-tuned models, hybrid retrieval, and local deployment options.

dhivyeshrk /Retrieval-Augmented-Generation-for-news

RAG system with real-time news scraping built using mixtral-8x7b, ChromaDB, bart summarizer

SciPhi-AI /r2r-js

js client for R2R: production-ready RAG engine with a sh*t ton of features.

vitrivr /vitrivr-engine

vitrivr's next-generation retrieval engine. It is capable of extracting and retrieving a wider range of multimedia objects such as audio, video, images or 3d models.

lucasdmarshall /Gyrfalcon

Hybrid retrieval for RAG in Rust. BM25 + dense search fused in one embedded engine. Same quality as LlamaIndex, LangChain and Haystack, 40x+ faster queries, 3 to 7x less memory. Python, Rust, CLI, HTTP.

vysakh0 /redhop

RedHop — a reasoning-aware retrieval & context runtime for RAG (in-process, no vector DB). Light weight alternative to langchain and llamaindex for retrieval alone.

alamgirqazi /DeReC

When retrieval outperforms generation: Dense evidence retrieval for scalable fake news detection - LDK 2025

EdoardoBotta /RecTokens

[Pytorch] Efficient tokenization for recommendations and generative retrieval. Inspired by STATIC decoding from "Vectorizing the Trie"