Library/LEANNForked

yichuan-w/LEANN

LEANN

RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.

Builder

yichuan-w

yichuan-w

yichuan-w • individual

Stars

10,387

Using upstream star count

Forks

900

Using upstream fork count

Open Issues

0

Activity Score

0/100

0 commits in 30d

Created

Jun 9, 2025

Project creation date

README Summary

LEANN is a privacy-focused RAG (Retrieval-Augmented Generation) application that runs entirely on personal devices without requiring cloud services. It achieves 97% storage savings through efficient compression techniques while maintaining fast and accurate performance for document retrieval and question answering.

AI Dev Skills

Unmapped

Retrieval-Augmented GenerationVector Database OptimizationDocument EmbeddingStorage CompressionOn-device ML DeploymentPrivacy-Preserving AIInformation RetrievalSemantic Search

Tags

Retrieval-Augmented GenerationVector Database OptimizationDocument EmbeddingStorage CompressionOn-device ML DeploymentPrivacy-Preserving AIInformation RetrievalSemantic SearchEdge/MobilePersonal Knowledge Base SearchTextLocal-first AIHealthcareLocal Information ExtractionPrivate Document Question AnsweringLegal TechEducationEfficient AI SystemsDocument ManagementSelf-hostedOffline Document RetrievalOn-device AIOn-premisePrivacy-Compliant RAG ApplicationsKnowledge ManagementEdge ComputingPython

Taxonomy

Recent Activity

Updated 5 months ago

7 Days

0

30 Days

0

90 Days

0

Quality

prototype
Quality
medium
Maturity
prototype

Categories

Inference & ServingPrimaryNLP & TextHealthcare & BiologyFinance & LegalEdge & Mobile AISearch & KnowledgeDev Tools & AutomationRAG & RetrievalEvals & BenchmarkingOther AI / ML

PM Skills

Developer Platform

Languages

Python100.0%

Timeline

Project created
Jun 9, 2025
Forked
Nov 8, 2025
Your last push
5 months ago
Upstream last push
9 days ago
Tracked since
Nov 8, 2025

Similar Repos

pgvector cosine similarity · $0

Loading…